A huge $12 billion complex to turn highly radioactive sludges into benign glass is the biggest project yet to clean up the radioactive wastes of Hanford nuclear reservation next to the Tri-Cities. But conflicting forces raise the question of whether the project is being done safely.
Two major considerations have been tugging the project in opposite directions — getting the project done on time versus making sure it will work safely. The people in charge of the project contend both concerns are being taken care of simultaneously. But internal critics, working for the same federal agency and same project contractor, say design problems are being deliberately and unsafely ignored to keep the project on schedule.
Since Hanford's cleanup efforts began in 1989, this tension has existed. The site frequently receives criticism for dawdling on cleanup; and the site frequently receives criticism for cutting corners in trying to keep to the latest schedule for the often-delayed cleanup.
Another factor may be in play. Hanford's critics point to the regular turnover in Hanford's top leaders in Richland and in Washington D.C. They argue that a top official might spend two or three years on a project and move elsewhere, just long enough to record doing good things on a resume without having to live with whether the glassification project will work as advertised when it goes online in 2019.
Hanford's master plan is to build a $12.3 billion complex to mix those wastes with liquid glass to create glass cylinders capable of holding in the radioactivity for 10,000 years. This troubled glassification project has a history of delays and cost increases. The price tag of $4 billion in 2002 grew $12.3 billion by early 2006 and has stayed steady since. The original start-up date of 1999 has been delayed to 2007, then to 2011 and now to 2019.
"Everyone is worried in the moment about their careers and jobs — and 2019 (when the glassification plant becomes active) is nine years away. Whoever you are talking to today, chances are they won't be there two years from now," said Tom Carpenter, director of Seattle-based watchdog organization Hanford Challenge in a late 2010 interview.
"It's the game of deny and delay," said Walt Tamosaitis, a veteran Hanford engineering team leader who used to be in charge of making sure a key component of the glassification plant would work as planned. The deadline to fix design problems in that key component, a pretreatment plant to prepares wastes for glassification, was June 30, 2010. the U.S. Department of Energy was supposed to pay Bechtel National Inc. and its lead subcontractor URS Corp. $5 million for meeting that deadline. At that time, DOE, Bechtel and URS agreed that the deadline was met, justifying the $5 million payment.
But in the months leading up to the deadline, Tamosaitis argued that the engineering problems were not adequately addressed. He still contends they have not been fixed. And on July 2, 2010, without warning, URS transferred Tamosaitis to a minor procurement job. Tamosaitis has filed a lawsuit in Benton County Superior Court against Bechtel and URS, alleging that he was transferred because he was raising legitimate concerns that would have ensured the June 30, 2010 would not be met. DOE and Bechtel say he was routinely transferred because his team's work was done.
Tamosaitis' lawsuit alleged that Russo and William Gay, URS's assistant manager for the glassification project, pushed hard to get an issue around the plant's mixing of the wastes resolved by June 30, 2010 in order to get the $5 million award fee. Tamosaitis charged that the pair pushed for weakening the standards which the mixing segment had to meet. Tamosaitis' lawsuit alleged that between February and June 2010, Gay repeatedly mentioned how not meeting the June 30 deadline could hurt careers and compensation. The lawsuit said: "On one or more occasions, Gay stated, 'If (the mixing issue ) doesn't close, I'll be selling Amway in Tijuana.' "
Over the last several months, DOE and Bechtel declined to let the appropriate officials to answer press questions, preferring to let spokespeople to address questions. The spokespeople said the design problems are either resolved or plans have been mapped out to tackle the unresolved engineering questions.
In the glassification complex, radioactive wastes will be pumped from the 177 tanks to the first stage of the glassification complex, which will be the 12-story, 540-foot-long, 215-foot-wide "pretreatment" building. There, the radioactive wastes will be mixed and separated into milkshake-like slurries to go to neighboring radioactive waste glassification buildlngs. Starting in 2019, almost all of the pretreatment building's insides will be too radioactive for people to enter.
Consequently, the pretreatment building's mixing tanks and pipes won't have moveable parts. Mixing will be done by air pumps that act like turkey basters in sucking air out of and blowing it into the tanks. If something has to be be replaced, an entire section of equipment will be removed by remote control on tracks to the building's huge central corridor — also too radioactive for people to enter.
Tamosaitis was in charge of making sure the 38 mixing tanks will work as desired. Issues included getting the solids small enough and spread out enough so they don't clog pipes or cause criticalities. All sides in this controversy agreed that criticalities are rare, but are huge problems when they occur. Meanwhile, the clogging can slow down how fast wastes are mixed, which could extend the amount of time that Hanford needs to glassify the wastes.
Other mixing issues include inevitable burps of hydrogen gas. How big will they be? Are explosions possible? Can they catch on fire? Will they bend the pipes and tanks enough that replacements will be needed? Will the chemicals react and bind as needed so the slurries can be later glassified? Will solids or sludges collect in the pretreatment plant's 32 tanks, possibly increasing the likelihood of a criticality occurring? Will the mixtures leave the pretreatment building in the proper chemical and physical conditions?
The stakes are high. If the mixing tanks don't work as designed, the pretreatment plant will have to be redesigned and rebuilt. That could translate to billions of extra dollars and some extra years before glassification could start again. But other stakes are involved. Falling behind the design and construction schedules lead to DOE paying millions of dollars less to Bechtel and URS.
When contacted recently, DOE spokeswoman Lori Gamache said: "The department has established strict nuclear safety criteria for the facility that must be met before any operations can begin. We will not allow the (tanks) to be installed until the results from large-scale testing have shown the the (tank) designs will safely and effectively handle the wastes at the site." Bechtel echoed that statement.
Tamosaitis contends the engineering questions are still unresolved. Several other experts, including Donald Alexander of the Department of Energy's nuclear safety division, agree with him, and the Defense Nuclear Safety Board has called for increased precautions around the transfer of materials to the glassification plant.
On July 1, 2010, the Consortium for Risk Evaluation and Stakeholder Participation — an independent technical review team under contract to DOE — released a report that found uncertainties with the pulse-jet mixer performances in the pretreatment plant and voiced several other safety concerns including the need for full-scale testing. Another concern was that different wastes in different mixtures exist in those of the 177 tanks that would be pumped into the pretreatment plant. And no one has a good handle on the exact chemical compositions in those tanks, and that knowledge could be vital for the pretreatment process to work.
One disputed issue is whether Hanford correctly extrapolated from a tabletop mixing test to a full-sized set of equipment. That dispute shows up in a May 20, 2010 email from David Dickey to Tamosaitis. Dickey is the head of MixTech Inc., an Ohio-based chemical engineering consulting firm that specializes in chemical mixing issues, and which was hired by the project's contractors. Dickey wrote: "The use of the 0.18 exponent for scale-up would be considered by me to be criminally negligent with respect to the design of a nuclear waste processing plant. Is this response worded strong enough?" When chemical engineers extrapolate from small-model tests to a full-sized pulse-jet mixing tank in the pretreatment plant, a scaling factor of 0.33 is an industry standard.
DOE and Bechtel spokespeople said that future tests are being planning to address scaling up the calculations from a tabletop size to a full-size mixing tank. Large-scale tests will be conducted before installing the tanks will be locked in, they said.