Technicians at the government’s Los Alamos National Laboratory settled on what seemed like a surefire way to win praise from their bosses in August 2011: In a hi-tech testing and manufacturing building pivotal to sustaining America’s nuclear arsenal, they gathered eight rods painstakingly crafted out of plutonium, and positioned them side-by-side on a table to photograph how nice they looked.

At many jobs, this would be innocent bragging. But plutonium is the unstable, radioactive, man-made fuel of a nuclear explosion, and it isn’t amenable to showboating. When too much is put in one place, it becomes “critical” and begins to fission uncontrollably, spontaneously sparking a nuclear chain reaction, which releases energy and generates a deadly burst of radiation.

The resulting blue glow — known as Cherenkov radiation — has accidentally and abruptly flashed at least 60 times since the dawn of the nuclear age, signaling an instantaneous nuclear charge and causing a total of 21 agonizing deaths. So keeping bits of plutonium far apart is one of the bedrock rules that those working on the nuclear arsenal are supposed to follow to prevent workplace accidents. It’s Physics 101 for nuclear scientists, but has sometimes been ignored at Los Alamos.

As luck had it that August day, a supervisor returned from her lunch break, noticed the dangerous configuration, and ordered a technician to move the rods apart. But in so doing, she violated safety rules calling for a swift evacuation of all personnel in “criticality” events, because bodies — and even hands — can reflect and slow the neutrons emitted by plutonium, increasing the likelihood of a nuclear chain reaction. A more senior lab official instead improperly decided that others in the room should keep working, according to a witness and an Energy Department report describing the incident.

Catastrophe was avoided and no announcement was made at the time about the near-miss — but officials internally described what happened as the most dangerous nuclear-related incident at that facility in years. It then set in motion a calamity of a different sort: Virtually all of the Los Alamos engineers tasked with keeping workers safe from criticality incidents decided to quit, having become frustrated by the sloppy work demonstrated by the 2011 event and what they considered the lab management’s callousness about nuclear risks and its desire to put its own profits above safety.

When this exodus was in turn noticed in Washington, officials there concluded the privately-run lab was not adequately protecting its workers from a radiation disaster. In 2013, they worked with the lab director to shut down its plutonium handling operations so the workforce could be retrained to meet modern safety standards.

Those efforts never fully succeeded, however, and so what was anticipated as a brief work stoppage has turned into a nearly four-year shutdown of portions of the huge laboratory building where the plutonium work is located, known as PF-4.

Officials privately say that the closure in turn undermined the nation’s ability to fabricate the cores of new nuclear weapons and obstructed key scientific examinations of existing weapons to ensure they still work. The exact cost to taxpayers of idling the facility is unclear, but an internal Los Alamos report estimated in 2013 that shutting down the lab where such work is conducted costs the government as much as $1.36 million a day in lost productivity.

And most remarkably, Los Alamos’s managers still have not figured out a way to fully meet the most elemental nuclear safety standards. When the Energy Department on Feb. 1 released its annual report card reviewing criticality risks at each of its 24 nuclear sites, ranging from research reactors to weapon labs, Los Alamos singularly did “not meet expectations.”

In fact, Los Alamos violated nuclear industry rules for guarding against a criticality accident three times more often last year than the Energy Department’s 23 other nuclear installations combined, that report said. Because of its shortcomings, federal permission has not been granted for renewed work with plutonium liquids, needed to purify plutonium taken from older warheads for reuse, normally a routine practice.

Moreover, a year-long investigation by the Center makes clear that pushing the rods too closely together in 2011 wasn’t the first time that Los Alamos workers had mishandled plutonium and risked deaths from an inadvertent burst of radiation. Between 2005 and 2016, the lab’s persistent and serious shortcomings in “criticality” safety have been criticized in more than 40 reports by government oversight agencies, teams of nuclear safety experts, and the lab’s own staff.

The technicians’ improvised photo-op, an internal Energy Department report concluded later, revealed the staff had become “de-sensitized” to the risk of a serious accident. Other reports have described flimsy workplace safety policies that repeatedly left workers uninformed of proper procedures and left plutonium packed hundreds of times into dangerously close quarters or without appropriate shielding to prevent a serious accident.

Workplace safety, many of the reports say, has frequently taken a back seat to profit-seeking at the Los Alamos, New Mexico, lab — which is run by a group of three private firms and the University of California — as managers there chase lucrative government bonuses tied to accomplishing specific goals for producing and recycling the plutonium parts of nuclear weapons.

And these safety challenges aren’t confined to Los Alamos. The Center’s probe revealed a frightening series of glaring worker safety risks, previously unpublicized accidents, and dangerously lax management practices. The investigation further revealed that the penalties imposed by the government on the private firms that make America’s nuclear weapons were typically just pinpricks, and that instead the firms annually were awarded large profits in the same years that major safety lapses occurred. Some were awarded new contracts despite repeated, avoidable accidents, including some that exposed workers to radiation.

Asked about this record, spokesman Gregory Wolf of the National Nuclear Security Administration (NNSA), which oversees and pays for the country’s nuclear weapons work, responded that “we expect our contractors to perform work in a safe and secure manner that protects our employees, our facilities, and the public. When accidents do occur, our focus is to determine causes, identify corrective actions, and prevent recurrences.”

His colleague James McConnell, the top NNSA safety official, said in an interview that “safety is an inherent part of everything we do.” But at a public hearing in Santa Fe on June 7, McConnell was also candid about Los Alamos’s failure to meet federal standards. “They’re not where we need them yet,” he said of the lab and its managers.

Los Alamos spokesman Kevin Roark said in an email the lab chose to defer to NNSA for its response. But the lab’s director over the past seven years, nuclear physicist Charles McMillan, said in a 2015 promotional video that while “we’ve got to do our mission” — which he said was vital to the nation’s security as well as the world’s stability — “the only way we can do that is by doing it safely.”

No usable warhead production for four years

The huge, 39-year-old, two-story, rectangular building at Los Alamos where the 2011 incident occurred is the sole U.S. site that makes plutonium cores — commonly known as pits because they are spherical and placed near the center of nuclear bombs — for the warheads meant to be installed over the next three decades in new U.S. missiles, bombers, and submarines.

Production of these cores is a key part of the country’s effort to modernize its nuclear arsenal at a cost of hundreds of billions of dollars, which President Obama supported and President Trump has said he wants to “greatly strengthen and expand.” Trump’s proposed fiscal year 2017 and 2018 budgets would boost U.S. spending on such work by $1.4 billion, representing a slightly higher percentage increase (11%) than requested overall for the Defense Department.

But mostly because of the Los Alamos lab’s safety deficiencies, it hasn’t produced a usable new warhead core in at least six years. Congress mandated in the 2015 National Defense Authorization Act that Los Alamos must be capable of manufacturing up to 20 war-ready cores a year by 2025, 30 the next year and 80 by 2027. Wolf said the agency remains committed to meeting this goal, but other government officials say the dramatic slowdown at PF-4 has put fulfillment of that timetable in doubt.

PF-4 is also the only place where existing cores removed randomly from the arsenal can be painstakingly tested to see if they remain safe and reliable for use in the nuclear stockpile. That work has also been blocked, due to PF-4’s extended shutdown, according to internal DOE reports.

The lab tried to conduct those tests in late 2016, but without success. The initial experiment destroyed a plutonium pit without collecting useful results about its safety or reliability, the latest annual review of Los Alamos’ performance by the National Nuclear Security Administration (NNSA) stated. The lab canceled a second planned pit analysis in 2016, according to the NNSA’s annual evaluation of the lab’s performance.

“I don't think they've made mission goals the last four years,” said Michaele Brady Raap, a past president of the American Nuclear Society and member of the Energy Department’s elite Criticality Safety Support Group, a team of 12 government experts that analyzes and recommends ways to improve struggling federal nuclear safety programs.

The lab’s criticality safety shortcomings have been so persistent that NNSA in August 2015 threatened to fine Los Alamos’ managing contractors more than a half-million dollars for failing to correct them. In the end, the NNSA administrator decided to not to impose the fine, exemplifying what critics allege is a climate of impunity for mistakes within DOE.

“There is no doubt, they have had some management and operational problems,” said MIT Professor Ernest Moniz, who served as the Obama administration’s Energy Secretary from 2013 until the end of January, speaking about Los Alamos’s handling of nuclear safety. “We were obviously quite concerned about it.”

Moniz said in an interview with the Center that the laboratory’s lapses had played a role in the department’s decision last year not to extend its existing management contract. Instead, the contract was opened to a new competition, with the winner expected to be named in early 2018 and take over the lab in Sept. 2018. Moniz added, however, that in 2016 the lab “started to turn things around.”

But others see Los Alamos’s conduct differently. “There’s a systemic issue here,” said Brady Raap. “There are a lot of things there [at Los Alamos] that are examples of what not to do.”

George Anastas, a past president of the Health Physics Society who analyzed dozens of internal government reports about criticality problems at Los Alamos for the Center, said he wonders if “the work at Los Alamos [can] be done somewhere else? Because it appears the safety culture, the safety leadership, has gone to hell in a handbasket.”

Anastas said the reports, spanning more than a decade, describe “a series of accidents waiting to happen.” The lab, he said, is “dodging so many bullets that it’s scary as hell.”

Ghastly deaths after the blue glow

The consequences of a “criticality” accident are ghastly. When Japanese technicians sloppily packed too much enriched uranium — another nuclear weapons fuel — into some wide-mouthed buckets at a factory 75 miles northeast of Tokyo in September 1999, it started to fission spontaneously in a classic “criticality” incident.

Two Japanese workers died, neighboring towns were contaminated with radiation, and industries essential to the region’s economy were disrupted. Schools closed, police barricaded roads, and trains stopped running. More than 160 people within a quarter-mile were evacuated, and another 310,000 people living and working nearby were ordered to seek shelter.

There was no explosion, just the usual blue Cherenkov flash, marking the spread of radiation around the Tokaimura plant in a chain reaction that pulsed intermittently for 20 hours. It exposed 119 people to doses exceeding the 1 millisievert level recognized by the International Commission on Radiological Protection as the maximum that members of the public can safely be exposed to in a year, according to the World Nuclear Association, a nonprofit organization that advocates expanded reliance on nuclear energy. Those contaminated were a mix of plant workers and others who by chance happened to live or work nearby.

Hisashi Ouchi and Masato Shinohara, who were in the room where the criticality occurred and absorbed extremely high doses — 1,700 and 1,200 rems of radiation, respectively — appeared normal when they entered the University of Tokyo Hospital Emergency Department on the same day. But within weeks, Ouchi became unrecognizable, inside and out.

Slowly, his skin sloughed off and his muscle tissue died. Externally, his body withered into a skeletal silhouette, covered in open sores. Inside his body, his chromosomes shattered like glass. Sequentially, his organs failed. By the 63rd day of his ordeal, doctors were pumping 10 liters of liquid into Ouchi to replace the fluid he was losing from surface wounds and massive intestinal bleeding. He died in December, 1999, 83 days after the accident.

Shinohara’s physical decline wasn’t as meticulously chronicled as Ouchi’s. But the outer layer of his skin molted from 70 percent of his body, and his body shut down in the same sequence that Ouchi’s had. He lived for 210 days after the accident, until he succumbed to MRS pneumonia on April 27, 2000.

Official studies by the Japanese government and the U.S. Nuclear Regulatory Commission chronicled a poor safety culture that had discounted the likelihood of a criticality accident. In 1999, Sixto T. Almodovar, a senior nuclear criticality analyst consultant at the Hanford Nuclear Site in Washington state, summarized the mindset about criticality safety at JCO Co. Ltd., the company that operated the Japanese nuclear fuel plant, as “Titanic thinking.”

“This ship is unsinkable, therefore why obstruct the view of the first-class passengers with unneeded life boats,” Almodovar said. Citing Japanese media reports, he noted that company officials had admitted they not only condoned, but encouraged, workers to take shortcuts, often at the expense of safety, to increase their productivity.

Taking safety shortcuts to boost productivity to the level managers wanted to see isn’t just a foreign problem, Almodovar warned. At the Y-12 National Security Complex, an Energy Department-funded nuclear weapons plant in Tennessee, workers even coined a euphemism for the practice. “The Oak Ridge Y-12 workers call this, a ‘Bubba said,’” Almodovar said, after interviewing some of them. A spokeswoman at Y-12, Ellen Boatner, didn’t reply to a request for comment.

Repeatedly playing with danger

Los Alamos’s first death from criticality-produced radiation occurred in September, 1945, 25 days after physicist Harry Daghlian deliberately lowered a large piece of plutonium into a cavity made of tungsten bricks that reflected the plutonium’s escaping neutrons back toward it, in a risky experiment that scientists dubbed “tickling the dragon’s tail.”

As Daghlian moved the final brick closer to the stack, a nearby radiation meter clicked frantically as neutrons collided angrily with other particles, warning him that a criticality accident loomed. But as he tried to withdraw the brick, it dropped, and the flash caught him. He died 28 days after he was irradiated.

The following May, Los Alamos scientist Louis Slotin was also testing the boundaries of plutonium criticality while seven other scientists looked on. Slotin was positioning a spherical beryllium shell around a plutonium pit. But as he slowly lowered the upper hemisphere onto the lower one, it slipped downward, off the tip of his screwdriver.

The telltale blue flash that followed gave Slotin enough radiation to kill him five times over, and the seven observers in the room received doses ranging from nearly lethal to benign. Slotin prevented a worse calamity by quickly separating the two halves of the pit, before the reaction could become self-sustaining. Nine days later, he died at the age of 35.

It happened again at Los Alamos, twelve years later, when chemist Cecil Kelley stood on a small ladder to stir a vat that included plutonium residue. When it became too concentrated, workers outside saw a bright blue flash and heard a dull thud. Soon, they saw Kelley standing outside, bewildered. “I’m burning up!” he screamed. “I’m burning up.”

The first medics to treat Kelley weren’t sure what had happened because he was working alone and too stunned to describe his experience. A nurse, among the first to treat him, didn’t suspect he’d been exposed to radiation and remarked on his “nice pink skin,” a sunburn-like symptom of his radiation exposure, according to an account of the accident  published in the journal Los Alamos Science in 1995.

Kelley died at the hospital in Los Alamos about 35 hours after the accident.

These deaths were all avoidable.  “The human element was not only present but the dominant cause in all of the accidents,” a team of criticality safety experts from Los Alamos and their Russian counterparts wrote in a definitive study of 60 criticality accidents published by the lab in 2000.

Reports over the past decade suggest, however, that these mistakes didn’t have a huge impact on criticality practices at Los Alamos. That lab has always been the most prominent and best funded — and according to Secretary of Energy Samuel Bodman’s notorious remark at a 2007 congressional hearing, the most infected by “arrogance” and resistant to independent scrutiny — of the U.S. nuclear weapons laboratories.

In 2005, shortly before the profit-making firms wrested majority control of the laboratory from the University of California, the lab’s “nuclear criticality safety program did not meet many of the” nuclear industry’s standards, according to a DOE report in 2008.

“We couldn’t prove we were safe,” said Doug Bowen, a nuclear engineer who was on the laboratory’s criticality safety staff at the time, “not even close.”

Two months after the takeover, the Defense Nuclear Facilities Safety Board — an independent federal oversight agency in Washington — concluded that the lab’s staff of 10 criticality safety engineers would need to more than triple. Its chairman also said the deficiencies hadn’t gotten adequate attention from the NNSA.

Los Alamos’s director of nuclear and high-hazard operations at the time, Robert McQuinn, dismissed that complaint in a written reply the following month. “LANL does not believe an inadvertent criticality is credible,” McQuinn said, without referencing the lab’s history. But he also promised the lab “has and is continuing to make significant progress in resolving the issues.”

But safety was not the foremost concern in Washington. To encourage higher efficiency and productivity, the Energy Department waved millions of dollars at its new corporate partners* as potential rewards for meeting deadlines for designing weapons and building bomb components at PF-4. Doing so created a mindset among managers and their work crews that posed challenges for safety experts like Bowen.

“Operations is always going to try to push the boundaries so they can produce as much as they can within the safety envelope when they’re pushing to get something done,” Bowen said. “Occasionally, they make decisions that they assume are going to be okay” but instead wind up exceeding limits, he explained.

A bonus was also offered if the laboratory started meeting basic criticality safety standards. But Bowen said that, in his view, meeting minimum requirements shouldn’t need and didn’t deserve bonus pay.

The new corporate group promised to bring the lab up to the required safety standards in 2007. But that September, when members of the Defense Nuclear Facilities Safety Board inspected plutonium vaults at PF-4, they discovered much more material present than inventories showed, posing new risks of spontaneous fissioning if some of it became too tightly packed together. So in September 2007, the lab shut down PF-4 for a month and told DOE it had created a Nuclear Criticality Safety Board to analyze and fix the lab’s persistent problems.

In 2010, when the Energy Department did a checkup, however, it found “no official notes or records” the group had ever met, according to an internal Energy Department report. The lab’s promised date to improve criticality safety had slid to 2008, then 2010, and then to 2011.

Too much plutonium in a glove box

When a nuclear technician put those eight plutonium rods dangerously close together on the afternoon of Aug. 11, 2011, he used a “glove box” — a device with gloved portholes that is designed to contain any radioactive particles — that he lacked permission to use. A sign on the box specifically warned against packing too much material inside, but he ignored it and went roughly 25 percent over the limit.

In one photo, obtained by the Center, two of the rods are touching each other as they rest on a roll of duct tape.  In another, eight rods are clustered tightly enough to fit within a pencil’s length, propped up against a pyramid-shaped stick with black and yellow candy stripes to indicate “caution.” Workers had forged the plutonium rods as aliquots — samples that could be useful to researchers in the weapons program and to teams trying to perfect the conversion of weaponized plutonium into fuel for civilian power plants.

Bowen, who was then Los Alamos’s top criticality safety expert and now supervises safety work throughout the weapons program, recalls getting a phone call about the technicians’ error from an assistant lab director around 90 minutes after it had been discovered. By then, the rods had already been picked up and moved by hand while other work in the room continued — contrary to procedures calling for an evacuation, his immediate notification, and for the dispatch of workers in hazmat suits to reconfigure the rods.

It was also a violation of the approach McMillan touts in the LANL promotional video. “I think it’s critical that if something doesn’t feel quite right, then you pause the work,” MacMillan said there. “It’s a lot better to stop than it is to just muscle through.”

Reaching into the box was dangerous, said Don Nichols, the NNSA associate administrator for safety and health at the time, because the water present in human bodies reflects neutrons and slows their speed, increasing the likelihood that those emitted by plutonium will collide with the nuclei of other plutonium atoms and emit more neutrons, triggering a nuclear chain reaction, with its accompanying release of energy and radiation.

As a result, Nichols said, the first thing to do upon noticing a near-criticality is “the opposite of what you want to do,” such as reach in and separate the offending materials. Instead, he said, those in charge should get “everyone to back off” and then call for engineers to start calculating safe approaches.

When Bowen reached the site, it was bathed in red lights as a belated warning for workers to stay away. He found the photographer looking despondent, with his head in his hands. Nearby, other workers consoled the equally upset technician. Both men were worried they’d be fired. During a lab-wide safety training a few days later, one of Los Alamos’ top safety officials called it “the most severe event” in years involving nuclear safety there, according to a copy of his presentation.

“The really horrible part that stuck in my mind is that they got lucky,” Bowen said. “They violated all these controls. They could have brought in more material to take pictures,” and had they done so, it could have cost the technicians their lives, never mind their jobs. Senior managers, he said, delayed calling in safety experts because they didn’t want to see the episode revealed in bold headlines.

“The management saw it as more of a political thing,” Bowen said. “They didn’t want this to get out in the papers or the news.” The fact that the call summoning him to PF-4 came from an assistant lab director — not a rank-and-file employee, but someone higher up — meant “they realized they were in trouble,” Bowen said.

The lab’s decision to downplay the risks of the 2011 incident was not an isolated one, Bowen added. An official with URS — one of the private contractors running PF-4 under a government contract — told Bowen “all the time that we don’t even need a criticality safety program,” Bowen recalled.

The URS official, Charles Anderson, who was appointed in July 2011 to oversee nuclear high-hazard safety, “basically said he didn’t need us and he could make more money” by replacing all the members of the criticality safety team with URS employees. (In 2014, a firm called AECOM acquired URS, including its stake in the consortium of contractors that operates Los Alamos.)

“That kind of culture really spawned the exodus” of the lab’s safety staff, Bowen said in an interview, which he gave to CPI before being promoted to his current leadership role in the NNSA criticality safety program. “Within a year, maybe a year and a half, there was one, maybe two left — 12 of 14 of the staff left. [And] because there was no criticality expertise there, it led to the closing of PF-4.”

It was, Bowen said, “a perfect storm of total boneheaded decisions by certain management [officials] at Los Alamos” that created such havoc. A former senior NNSA official in Washington recounted hearing a similar depiction of the URS contractor’s disdainful attitudes about criticality.

Numerous messages left on Anderson’s work and personal phones and emails as well as his social media accounts seeking comment went unanswered. A spokesman for the consortium of contractors that operates Los Alamos referred questions about Anderson’s reported actions to the NNSA, whose spokesman didn’t address those specific questions. AECOM, which bought URS, also did not reply to request for comment.

A special expert group created to monitor safety throughout the Energy Department’s facilities, known as the Criticality Safety Support Group, documented the exodus of trained personnel from Los Alamos in an April 2012 report, which said that experts “had lost trust in their line management.”

Nichols recalled in an interview that due to “some mismanagement, people voted with their feet. They left.” The attrition rate was around “100%,” according to a private “lessons learned” report last month by the lab’s top criticality expert and the lone NNSA criticality expert assigned to work there, which they prepared for counterparts at the nearby Sandia nuclear weapons lab.

The 2011 incident “was an egregious event,” agreed Brady Raap, who has been a chief engineer in the nuclear engineering and analysis department of the national security division at the Energy Department’s Pacific Northwest National Laboratory. “That was what said, really, ‘Look, there’s not the respect for safety that there needs to be.’ The problem was more than a few disgruntled people or anything that people [in management] portrayed it as.”

“Operations wasn’t fully integrated with safety, as it should be,” she said. “There’s an inherent conflict between safety objectives, which can slow down work, and productivity pressure…. Management, in particular, is focused on a mission goal — processing a certain amount of material or manufacturing enough widgets, or what have you. If they don’t have enough respect for the safety activities that support that, things become a little detached. You proceed when it would have been better to wait.”