# The 2 Billion Dollar Cat Litter: How One Nuclear-Waste Drum Shut the Only US Repository

About 2 billion dollars: what one drum of nuclear waste cost after the wrong, wheat-based cat litter reacted underground and shut the only US repository for years.

Author: J.A. Watte
Published: July 16, 2026
Source: https://jwatte.com/blog/wipp-nuclear-waste-accident/

---


There is a place in the New Mexico desert, half a mile below the surface, where the United States buries the most dangerous garbage of its nuclear-weapons program. It is the only deep geologic repository the country has ever managed to open and operate, the one piece of American nuclear-waste policy that actually got built. And on a February night in 2014 it was taken offline for roughly three years by a single 55-gallon drum, packed a thousand miles away, whose contents had been mixed with the wrong kind of cat litter. The all-in cost of that mistake has been estimated at around 2 billion dollars. What makes the story worth telling is not that a mundane substitution caused a radiological accident, striking as that is, but that the facility it crippled is the success story of the program, not the failure. To understand the boondoggle you have to hold both facts at once.

## What WIPP is, and who actually runs it

The Waste Isolation Pilot Plant, universally called WIPP, sits about 26 miles east of Carlsbad, New Mexico. Its disposal rooms are mined roughly 2,150 feet underground in the Salado formation, a bed of Permian-age salt that has been geologically stable for more than 200 million years and that slowly creeps inward to entomb whatever is placed in it. It is run by the Department of Energy through a management-and-operations contractor. It opened for waste in March 1999 and remains, [as its own history records](https://www.sandia.gov/labnews/2024/10/31/the-history-of-wipp/), the only operating deep geologic repository in the United States.

The distinction that matters most here, and the one most easily blurred, is what WIPP is allowed to bury. It takes defense-generated transuranic waste, known as TRU: the gloves, tools, rags, machine parts, and laboratory debris contaminated with plutonium and other man-made heavy elements over decades of building nuclear weapons. It does not take commercial spent nuclear fuel, the intensely radioactive rods that come out of power reactors. That was supposed to be a different repository's job. [As the site's technical documentation makes plain](https://en.wikipedia.org/wiki/Waste_Isolation_Pilot_Plant), WIPP is a weapons-legacy facility, not a civilian-power one. If you take away one fact, take away that one, because the contrast it sets up is the whole point.

## The repository that opened, against the one that never did

For most of the past half-century, when Americans have argued about where to put nuclear waste, they have argued about Yucca Mountain, a ridge in the Nevada desert designated as the national repository for commercial spent fuel and high-level waste. Yucca consumed decades of study, tens of billions of dollars in planning and litigation, and enormous political capital. It never opened. Federal funding for the project was zeroed out in the 2011 budget, and while it has never been formally cancelled, it has sat unfunded and defunct ever since, [as the record shows](https://en.wikipedia.org/wiki/Waste_Isolation_Pilot_Plant).

So the country's two flagship attempts at permanent nuclear disposal split cleanly. Yucca, the famous one, is the repository that never received a single canister. WIPP, the obscure one, is the repository that quietly opened in 1999 and has been taking waste ever since. When you read what follows, keep that asymmetry in view. The accident that cost around 2 billion dollars happened at the facility that works. The far larger sums swallowed by Yucca bought a repository that does not exist.

## What went wrong, and the chemistry underneath it

At about 11:14 p.m. on February 14, 2014, a single drum of TRU waste breached underground at WIPP. Heat had built inside it until the drum over-pressurized, popped its lid, and vented. It released americium and plutonium contamination into the mine, and through a leaking ventilation damper that let a small fraction slip past the HEPA filters, a trace amount reached the surface environment. [The Environmental Protection Agency's account](https://www.epa.gov/radiation/2014-radiological-event-wipp) traces the release to one container, Los Alamos drum 68660. One drum, out of the tens of thousands emplaced there, did all of this.

The cause was chemistry, and it is worth getting right rather than settling for the tabloid shorthand. The drum held remediated nitrate salt waste. Nitrate salts are oxidizers: chemically, they are eager to give up oxygen and drive combustion. To stabilize such waste for packaging, handlers are supposed to add an inorganic absorbent, a clay or zeolite material that is chemically inert and simply soaks up liquid. Instead, this drum had been packed with an organic, wheat-based absorbent, a brand of cat litter called Swheat Scoop, substituted for the inorganic kind. Organic wheat cellulose is fuel. Put an oxidizer and a fuel in a sealed drum together, add the mildly acidic and reactive chemistry already present in the waste, and you have the ingredients for a slow exothermic reaction, one that generates its own heat and can run away thermally until something bursts. [Chemical & Engineering News walked through the mechanism](https://cen.acs.org/articles/95/i20/wrong-cat-litter-took-down.html): laboratory thermal-analysis experiments confirmed that nitrate salt, Swheat, and the drum's acids self-heat at temperatures below the boiling point of water, and the modeling matched the roughly 70-day interval between when the drum was packaged, in December 2013, and when it breached. The cat litter itself was never radioactive. It was the fuel in a reaction that the required inert absorbent would never have supported.

The error did not originate at WIPP. It originated at Los Alamos National Laboratory, where a change to the glovebox waste-packaging procedure specified an organic absorbent where an inorganic sorbent was intended. [A Department of Energy inspector-general review](https://cen.acs.org/articles/95/i20/wrong-cat-litter-took-down.html) pointed to exactly that procedure change. Los Alamos produced a noncompliant, ignitable waste form and shipped it to WIPP, which received it and buried it. WIPP was the site where the drum went off, but the mistake was born upstream, in a lab a thousand miles away, and traveled the length of the weapons complex without anyone catching the mismatch.

## The people, and why this was not a mass-casualty event

It is easy, with a phrase like "plutonium release," to imagine catastrophe. The measured record does not support that framing, and honesty requires stating the doses plainly. As of the March 28, 2014 bioassay data, 21 surface workers tested positive for low-level internal contamination. Their doses were reported well within safety limits, with no acute radiation harm to anyone. [The dose reconstruction](https://en.wikipedia.org/wiki/Waste_Isolation_Pilot_Plant) put the exposures at levels comparable to routine low-level contamination, each far below thresholds of concern. Some Department of Energy passages cite 22 rather than 21 as the total who received some internal contamination, each estimated at under 10 millirem of lifetime dose spread over 50 years; the figures are close and both are defensible, and the point they share is the important one. Off the site, the calculated public dose was less than 0.01 millisievert per year, about 1 millirem, comfortably beneath the 10-millirem annual limit.

This was, in other words, a serious operational and financial failure and a minor human-health one. Both halves are true. A facility whose entire justification is the disciplined, permanent isolation of the nation's most hazardous defense waste suffered a containment breach it was designed never to have. And the health consequence was 21 workers with low doses and no acute harm. Keeping those two facts in the same paragraph is what separates an accounting from a scare.

## What the investigation found

The Department of Energy did not treat this as a freak event, and neither should anyone reading about it. It convened an Accident Investigation Board, which issued a Phase 1 report on April 22, 2014, covering the release and the emergency response, and a Phase 2 report on April 16, 2015, submitted to the appointing official at the end of March, that identified the exothermic organic-nitrate reaction as the cause. [The Board's landing page and reports](https://www.energy.gov/ehss/articles/accident-investigations-february-14-2014-radiological-release-waste-isolation-pilot) describe not a lone bad actor but a chain of systemic failures: a degraded safety culture, inadequate review of chemical hazards, weak hazard controls, and thin federal oversight spread across the laboratory, the contractor, and the Department itself. There was, in the memorable summary of the findings, plenty of blame to go around.

That is the part that turns a chemistry mishap into a cautionary tale about public money. A single wrong absorbent in a single drum should have been caught by any one of several layers: the packaging procedure at Los Alamos, the chemical-compatibility review, the waste-acceptance screening at WIPP. It was caught by none of them, because the system built to catch exactly this kind of error had quietly eroded. The drum was the trigger. The eroded safety apparatus was the loaded weapon. (Note that a separate underground fire nine days earlier, when a salt-haul truck caught fire on February 5, 2014, was a genuinely unrelated event; it damaged the mine but had nothing to do with the radiological release, and the two should not be merged.)

## Why it looks like a loss

Now the money. The direct cost to recover the mine and reopen the facility was reported at roughly 500 million dollars, [a figure covered in the reopening reporting](https://www.forbes.com/sites/jamesconca/2017/01/10/wipp-nuclear-waste-repository-reopens-for-business/) and drawn from local accounting in the [Santa Fe New Mexican](https://www.santafenewmexican.com/news/local_news/cost-of-reviving-wipp-after-leak-could-top-million/article_1402a1fd-9e58-52c6-97df-0241375e839d.html). That is the narrow number: what it took to clean the contaminated mine, rebuild the ventilation, and turn the lights back on.

The broader number is larger, and it needs its label attached firmly. A Los Angeles Times analysis, reported through [The Ecologist](https://theecologist.org/2016/sep/20/wipp-nuclear-waste-accident-will-cost-us-taxpayers-2-billion), estimated the total cost of the accident and its aftermath at roughly 2 billion dollars. That figure combined about 640 million dollars in direct cleanup, tied to a contract modification with the site operator, and roughly 1.4 billion dollars in indirect and lost-operations costs, chiefly the extra years of running the facility at around 200 million dollars a year because full operations had been pushed out. This is an analyst's estimate with a broader scope than the 500-million reopening figure, not a single booked line in a federal ledger, and it should be read that way. It also, by its authors' own account, leaves some costs out, including full ventilation replacement and the expense of waste that sat stranded at other sites with nowhere to go while WIPP was closed. Under cost-plus contracting, the true all-in number could run higher still. The honest way to state it is a range: about half a billion dollars to reopen, on the order of 2 billion dollars all-in by the most-cited outside estimate.

There was a further bill. In January 2016, on the 22nd, the Department of Energy signed a settlement with the state of New Mexico worth about 74 million dollars over permit violations tied to the radiological release and the earlier truck fire, [described in the settlement coverage](https://www.foxnews.com/us/us-new-mexico-ink-74-million-in-settlements-over-nuclear-radiation-leak) as the largest ever between a state and the Department. Most of it was structured not as a cash penalty but as in-kind spending on New Mexico infrastructure, including upgrades to the transport routes the Department itself uses to move waste.

Set against the mission WIPP exists to perform, all of this reads as loss. The facility's single value proposition is that it handles the country's most dangerous defense waste with more care than anyone else. A procurement-and-paperwork error, an absorbent swapped for the wrong absorbent, that costs somewhere between half a billion and 2 billion dollars and idles the only working repository for three years is close to a textbook illustration of a small process failure metastasizing because the checks meant to contain it had gone slack.

## Ratio-adjusted for the mission it actually served

Change the measure, though, and the ledger does not simply reverse, but it does grow far more complicated, because the thing that broke is the thing that works.

WIPP is the one piece of American nuclear-waste policy that got built and opened and did its job. Since 1999 it has permanently removed plutonium-contaminated debris from decades of weapons work out of aging aboveground storage at sites scattered across the country and locked it 2,150 feet down in salt that will slowly seal around it for geologic time. That is not a hypothetical. It is the counterexample to Yucca Mountain, which absorbed vastly more money and political energy and never accepted an ounce of waste. Measured against the repository program's central and repeated failure, a facility that opened, suffered a serious accident, and came back is not the boondoggle of the story. It is the survivor.

And it did come back. The Department treated the accident as the systemic failure the investigation said it was. It rebuilt the contaminated ventilation system, overhauled waste-acceptance and safety-culture practices across the complex, and brought the repository formally back online on January 9, 2017, with waste shipments resuming that spring, [as the reopening reporting documented](https://www.forbes.com/sites/jamesconca/2017/01/10/wipp-nuclear-waste-repository-reopens-for-business/). The roughly three-year closure was real and expensive. So was the recovery. The mission the facility exists to serve, a working, geologically sound home for the weapons complex's most hazardous legacy, survived the accident and resumed, which is more than the rest of the repository program can say.

The human cost, held to the numbers, stays small. Twenty-one workers with low doses and no acute harm; a public dose off the site of under 1 millirem, below the regulatory limit. That is not nothing, and the safety-culture failures behind it are damning. But it is the difference between an accident that cost a great deal of money and a genuine disaster, and the record puts this squarely in the first category.

## The ledger reading

The WIPP accident is one of those episodes that is honestly described as an expensive failure and honestly described as a system that ultimately held, because the two verdicts answer two different questions. Measured as a discrete event, one wrong absorbent in one drum, traced to a slack safety apparatus, that cost on the order of a billion-plus dollars and idled a national asset for three years, it is indefensible, and the Department's own investigators said as much. Measured against the program it belongs to, WIPP is the repository that opened when the celebrated one never did, took a hard hit, and returned to service, which is the outcome the entire enterprise of nuclear-waste disposal has otherwise failed to produce.

What ties the two readings together is the same lesson that shows up wherever public money meets a hazard that must be managed forever. The value of a facility like WIPP is not in any single year's throughput; it is in the promise that the material stays put and the handling stays disciplined across decades. A promise like that is only ever as strong as its weakest check, and the accident is the proof: the redundancy was supposed to be many-layered, and a wheat-based cat litter walked through every layer. The 2 billion dollars, if that estimate holds, did not buy a cleanup so much as it bought a demonstration that a repository built to last ten thousand years can be undone for three by a procurement note. WIPP passed the test of coming back. It failed the test of not needing to.

## Related reading

- [Yucca Mountain and the Nuclear Waste Fund](/blog/nuclear-waste-fund-yucca-mountain/): the repository that collected tens of billions of dollars and never opened, the direct counterweight to the one that did.
- [The MOX Plutonium Fuel Facility](/blog/mox-plutonium-fuel-facility/): another weapons-legacy nuclear project measured by what it cost against what it delivered.
- [The Office of Secure Transportation](/blog/office-of-secure-transportation/): the federal operation that physically moves defense nuclear material, including the shipments that feed WIPP.

## Fact-check notes and sources

- **What WIPP is, where it sits, and what it holds** (a deep geologic repository about 26 miles east of Carlsbad, New Mexico, with rooms roughly 2,150 feet down in Salado salt beds; opened March 1999; the only operating US deep geologic repository; disposes of defense transuranic waste, not commercial spent fuel): the [Waste Isolation Pilot Plant technical record](https://en.wikipedia.org/wiki/Waste_Isolation_Pilot_Plant) (reputable secondary, corroborated by the Department of Energy) and [Sandia National Laboratories' WIPP history](https://www.sandia.gov/labnews/2024/10/31/the-history-of-wipp/).
- **The Yucca Mountain contrast** (planned commercial-spent-fuel repository; federal funding terminated in 2011; unfunded and defunct, though never formally cancelled): the same [WIPP technical record](https://en.wikipedia.org/wiki/Waste_Isolation_Pilot_Plant).
- **The accident and its cause** (single drum, Los Alamos 68660, breached about 11:14 p.m. on February 14, 2014, releasing americium and plutonium with a trace reaching the surface past HEPA filters via a leaking damper; caused by an exothermic reaction between nitrate salts and an organic wheat-based absorbent substituted for the required inorganic sorbent): the [Environmental Protection Agency's account of the 2014 radiological event](https://www.epa.gov/radiation/2014-radiological-event-wipp) and, for the chemistry, procedure change, and roughly 70-day birth-to-breach interval, [Chemical & Engineering News](https://cen.acs.org/articles/95/i20/wrong-cat-litter-took-down.html) (American Chemical Society).
- **Worker and public doses** (21 workers positive by bioassay as of March 28, 2014, with low doses and no acute harm; some Department passages cite 22 for total internal contamination; off-site public dose under 0.01 millisievert per year, below the 10-millirem limit): the [WIPP dose record](https://en.wikipedia.org/wiki/Waste_Isolation_Pilot_Plant). The 21-versus-22 discrepancy is noted in the source material; both figures describe low, non-acute exposures.
- **The investigation** (Accident Investigation Board Phase 1 report April 22, 2014, and Phase 2 report April 16, 2015, identifying the exothermic reaction and finding broad safety-culture and oversight failures across the lab, the contractor, and the Department): the [Department of Energy Office of Environment, Health, Safety and Security](https://www.energy.gov/ehss/articles/accident-investigations-february-14-2014-radiological-release-waste-isolation-pilot).
- **Closure and reopening** (roughly three-year closure; formal reopening January 9, 2017; shipments resuming around April 2017): [Forbes reporting on the reopening](https://www.forbes.com/sites/jamesconca/2017/01/10/wipp-nuclear-waste-repository-reopens-for-business/).
- **Costs** (about 500 million dollars for the narrow recovery-and-reopening scope; roughly 2 billion dollars all-in as a Los Angeles Times analyst estimate, combining about 640 million dollars in direct cleanup with about 1.4 billion dollars in indirect and lost-operations costs at roughly 200 million dollars a year; the about-74-million-dollar New Mexico settlement signed January 22, 2016, structured largely as in-kind infrastructure spending): the narrow figure from the [Santa Fe New Mexican](https://www.santafenewmexican.com/news/local_news/cost-of-reviving-wipp-after-leak-could-top-million/article_1402a1fd-9e58-52c6-97df-0241375e839d.html) and [Forbes](https://www.forbes.com/sites/jamesconca/2017/01/10/wipp-nuclear-waste-repository-reopens-for-business/); the all-in estimate and the settlement from [The Ecologist, citing the Los Angeles Times analysis](https://theecologist.org/2016/sep/20/wipp-nuclear-waste-accident-will-cost-us-taxpayers-2-billion), and the settlement date from [Fox News](https://www.foxnews.com/us/us-new-mexico-ink-74-million-in-settlements-over-nuclear-radiation-leak). The 2-billion-dollar figure is a media analyst estimate with a broader scope than the reopening cost and is labeled as such throughout; it is not a single booked federal line item.

*This post is informational and journalistic, describing a public program and public records. It is not legal, financial, or policy advice. Cost figures are drawn from government reports and news analysis; the roughly 2 billion dollar all-in total is an outside analyst's estimate, labeled as such, and is distinct in scope from the roughly 500 million dollar reopening cost. Dose figures are stated as reported by the Department of Energy and EPA.*

---

Canonical HTML: https://jwatte.com/blog/wipp-nuclear-waste-accident/
RSS: https://jwatte.com/feed.xml
JSON Feed: https://jwatte.com/feed.json
Hero image: https://jwatte.com/images/wipp-nuclear-waste-accident.webp