Whither Willie Pete?
Whither Willie Pete?
Originally shared by Yonatan Zunger
This waste site near Butte, Montana – a former phosphorus plant – hides an incredibly alarming thing: a 500,000-gallon concrete tank of white phosphorus sludge. That sludge sits under a cap of 2-3 feet of water, steadily replenished by an automated system; on top of that are rows of plastic "bird balls" that keep waterfowl from landing on the surface and slow evaporation. If that water cap were ever to fail, the phosphorus would come into contact with atmospheric oxygen, bursting instantly into flame. That would, in turn, clear out any remaining water, causing the entire pool to explode and strew toxic smoke across the entire area, with quite a few towns in the crossfire. For those of you familiar with them, that pool can basically be thought of as about 853,000 WP grenades with a water pumping system keeping their fuse unlit.
There are quite a few things wrong with this, to say the least. One is the economic chicanery which allowed it to come into existence in the first place: the company which operated the plant (Solvay) was essentially allowed to continue to accumulate risk while they operated, and then simply walk away when they were done, taking all the profits of having accumulated that risk and leaving the costs behind for anyone in the vicinity. This kind of negative externality is at the heart of most major waste sites. Fortunately, the EPA is holding Solvay's feet to the fire to make them deal with this mess – by Solvay's proposal, by building a reclamation plant to actually mine the useful phosphorus out of the pit, and from anyone else's phosphorus waste that they care to get rid of. (Which, if it can be pulled off, seems like a fairly good idea to me.) In many cases this isn't possible, since either the company which operated the things creating the waste sites simply no longer exists, or it operated the sites through a shell company that was later destroyed, specifically so that there would be no legally responsible party left around.
But there's a second very serious problem here, which isn't economics, it's engineering. The water cap which keeps this pit from exploding needs to be continually refreshed by a pumping system, with a connection to an active water supply, and so on. Fortunately, water levels normally decline slowly (mostly due to evaporation), and so even if the pumping system failed for some reason, humans could be alerted and begin emergency measures in time. But if anything interfered with that, or if the site were left derelict, or if something were to physically disturb the water cap (e.g., something big slamming into it and splashing the water out of the way)... boom.
The technical term for this problem is that it doesn't fail safe: it requires continuous active measures to keep it from exploding. In a properly designed system, the complete failure of all external support (or any other easily-predictable problem) should cause the system to drop itself into a safe, if not necessarily good, state.
It isn't always possible to make a system fail safe for physics reasons; for example, if you fully stop a nuclear reactor by dropping the control rods in all the way, the core is still physically very hot – more than hot enough to melt into rubble if the cooling system were to fail. That means there's no immediate way to bring a reactor to a "cold stop," the state where you can simply walk away from it and it's safe. To compensate for this, reactors have all sorts of mechanisms to keep the cooling systems working under a wide range of circumstances, as well as designs to ensure that even if the core does melt, that won't lead to a release of radioactives beyond the containment vessel. This approach is called "defense in depth," and it's crucial to any kind of safety system, not just a nuclear one.
If you're interested in this, I highly recommend James Mahaffey's Atomic Accidents (recommended to me a few years ago by Lea Kissner), a catalogue of every known accident in the history of nuclear physics, with a discussion of what went wrong and why. It shows how systems can be designed both well and poorly for worst-case disasters. (And will, I suspect, greatly increase your confidence in nuclear reactors.)
h/t @bridgietherease on Twitter.
http://mtstandard.com/natural-resources/the-waste-site-that-time-forgot-phosphorus-plant-s-fate/article_1ed352b2-752c-5a23-847d-d926b8e4ba41.html
Originally shared by Yonatan Zunger
This waste site near Butte, Montana – a former phosphorus plant – hides an incredibly alarming thing: a 500,000-gallon concrete tank of white phosphorus sludge. That sludge sits under a cap of 2-3 feet of water, steadily replenished by an automated system; on top of that are rows of plastic "bird balls" that keep waterfowl from landing on the surface and slow evaporation. If that water cap were ever to fail, the phosphorus would come into contact with atmospheric oxygen, bursting instantly into flame. That would, in turn, clear out any remaining water, causing the entire pool to explode and strew toxic smoke across the entire area, with quite a few towns in the crossfire. For those of you familiar with them, that pool can basically be thought of as about 853,000 WP grenades with a water pumping system keeping their fuse unlit.
There are quite a few things wrong with this, to say the least. One is the economic chicanery which allowed it to come into existence in the first place: the company which operated the plant (Solvay) was essentially allowed to continue to accumulate risk while they operated, and then simply walk away when they were done, taking all the profits of having accumulated that risk and leaving the costs behind for anyone in the vicinity. This kind of negative externality is at the heart of most major waste sites. Fortunately, the EPA is holding Solvay's feet to the fire to make them deal with this mess – by Solvay's proposal, by building a reclamation plant to actually mine the useful phosphorus out of the pit, and from anyone else's phosphorus waste that they care to get rid of. (Which, if it can be pulled off, seems like a fairly good idea to me.) In many cases this isn't possible, since either the company which operated the things creating the waste sites simply no longer exists, or it operated the sites through a shell company that was later destroyed, specifically so that there would be no legally responsible party left around.
But there's a second very serious problem here, which isn't economics, it's engineering. The water cap which keeps this pit from exploding needs to be continually refreshed by a pumping system, with a connection to an active water supply, and so on. Fortunately, water levels normally decline slowly (mostly due to evaporation), and so even if the pumping system failed for some reason, humans could be alerted and begin emergency measures in time. But if anything interfered with that, or if the site were left derelict, or if something were to physically disturb the water cap (e.g., something big slamming into it and splashing the water out of the way)... boom.
The technical term for this problem is that it doesn't fail safe: it requires continuous active measures to keep it from exploding. In a properly designed system, the complete failure of all external support (or any other easily-predictable problem) should cause the system to drop itself into a safe, if not necessarily good, state.
It isn't always possible to make a system fail safe for physics reasons; for example, if you fully stop a nuclear reactor by dropping the control rods in all the way, the core is still physically very hot – more than hot enough to melt into rubble if the cooling system were to fail. That means there's no immediate way to bring a reactor to a "cold stop," the state where you can simply walk away from it and it's safe. To compensate for this, reactors have all sorts of mechanisms to keep the cooling systems working under a wide range of circumstances, as well as designs to ensure that even if the core does melt, that won't lead to a release of radioactives beyond the containment vessel. This approach is called "defense in depth," and it's crucial to any kind of safety system, not just a nuclear one.
If you're interested in this, I highly recommend James Mahaffey's Atomic Accidents (recommended to me a few years ago by Lea Kissner), a catalogue of every known accident in the history of nuclear physics, with a discussion of what went wrong and why. It shows how systems can be designed both well and poorly for worst-case disasters. (And will, I suspect, greatly increase your confidence in nuclear reactors.)
h/t @bridgietherease on Twitter.
http://mtstandard.com/natural-resources/the-waste-site-that-time-forgot-phosphorus-plant-s-fate/article_1ed352b2-752c-5a23-847d-d926b8e4ba41.html
Comments
Post a Comment