Slashdot Mirror
New Neutron Scatter Camera to Detect Smuggled Nukes
Researchers at Sandia National Laboratories in California are developing a new neutron scatter camera that they claim will be able to detect radiation through much more shielding and at much greater distances than traditional tech. "The neutron scatter camera consists of elements containing proton-rich liquid scintillators in two planes. As neutrons travel through the scintillator, they bounce off protons like billiard balls. This is where "scatter" comes into play -- with interactions in each plane of detector elements, the instrument can determine the direction of the radioactive source from which the neutron came. [...] Computers record data from the neutron scatter camera, and using kinematics, determine the energy of the incoming neutron and its direction. Pulse shape discrimination is employed to distinguish between neutrons and gamma rays."
This is part of the problem with working under government contracts on projects and interests that are ultimately unrelated - you always have to justify the research. I worked in this department at Sandia for a few summers, and that was generally the opinion I heard. Everyone there is interested in fusion research to provide cheap power to the world, but they have to do this research under the auspices of supporting national security or nuclear stockpile stewardship. Everyone working there knows that that's not the actual reason for the research, but it keeps the funds flowing in.
1. DoD contractor produces a prototype, then obtains a $100M grant from the DoD to pursue it further.
2. DoD contractor requests $50M for additional research and receives it
3. DoD contractor delivers the detector as a proprietary black box, running Windows, at a price of $10M each. 50 units are ordered by the government.
4. 5 CalTech students make a working detector for $20'000 out of an old scintillation counter, plumbing pipe, and a PentiumIII machine running BSD.
5. Nobody cares.
It is funny how much of the federally funded research right now appears to be directly applied to counter-terrorism efforts. I am all for that being done, but from what I have noticed when going through the national labs web-pages, it seems that the majority of research dollars are going to these efforts. I think that if you counted military research labs, you might find that more federal dollars are going to counter-terrorism than are going to alternative energy projects.
To clarify, this neutron camera is nice for nukes, but what they're actually using the technology for is to examine neutron emissions from fusion capsules compressed with their z-pinch machine.
The researchers in this case are interested only in the science, but in order to get the government funding, they need to think up ways that their research could be used by the government. A hot-button issue for the government is national security, so applications related to that are the best to mention. In reality this neutron camera is for fusion research that could ultimately provide cheap power to the world.
I wish nobody cared, but this will only fuel the masses into thinking "wow... all this money spent on high-tech, super sci-fi counteterorism stuff is making me safer."
Don't get me wrong being able to detect a nuke is a good thing. However, to me this seems to fit right in along with the whole security theater schtick that the government is pulling. Throw out some nifty vaporware. Have some conveniently thwarted plots and you have a carte blanche to do whatever you want with personal liberty.
Without getting into an argument over the actual probability (well over-blown if you ask me) of a terrorist attack occurring, being nuked is the least of my fears. Something that utilizes few resources, like say hijacking a plane, is much more likely to be the plan of attack. The government throws out all these crazy, high-concept plots just to proliferate fear.
I know I'm off-topic for this specific device, but I feel that the whole counterterorism deal is what's behind this. As another post pointed out, it's what is driving research funding, and that makes me just a little uncomfortable.
I got a catholic block.
It has the "potential to detect through various types of shielding" and "through more shielding" (FTA) but what types of shielding and at what thicknesses? Soil, Rock, Water, Lead, Etc?...
This Reads: 'fancy new-fangled oceanic Shipping Container Nuke Detector' all over it, and maybe something new for surveillance aircraft too.
That kind of hot geek talk gets my protien rich liquid scintillator scintillated.
A local nuclear physicist in our area recently commented about the current detection systems in use. He regarded them as so easily thwarted they are basically worthless. He even described how to do it on a local news broadcast (sorry, I couldn't find a link). It basically involved very low cost (common) materials. He indicated the type of technology talked about in this article is really the only meaningful method for detecting nuclear material. He further stated that the organizations responsible for detecting this material know what they are using is worthless, but are unwilling to spend the additional money needed for the correct technology. He was upset that they were more interested in putting on a 'show' of force, rather than offering real protection.
Let's hope that isn't true, and places like Sandia are working on making nuetron detection less expensive.
Nick (assuming it's the same guy and not some other Nick Mascarenas) was a post-doc in the same lab as me at Caltech in the early '90's. We were working on a reactor neutrino experiment (now defunct) looking for neutrino oscillations. Discriminating against fast neutron backgrounds was an important part of the design problem.
What has been done here is fairly clever, although I'm doubtful as to the ultimate viability due to low cross-sections and high backgrounds and easy work-arounds by the bad guys.
Spontaneous fission produces fast neutrons, which are relatively hard to shield against. First they have to thermalize, then get captured. Things that are good at shielding gamma rays (heavy elements) are lousy at thermalizing neutrons (light elements), so it makes the bad guy's shielding problems harder to solve.
Ergo, if you can detect fast neutrons, and determine where they are coming from, you have a backup bomb detector that is harder to beat. The way Nick is proposing to do this is with a setup in which you have two planar liquid scintillator detectors and look for coincidences (suitably delayed by the neutron's quite significant travel time) between them. Fast neutrons deposit energy into the detectors via proton recoil, which creates a distinct kind of optical event from electron-positron showers produced by gamma rays. Furthermore, you tend to get forward scattering, so you can at least tell which hemisphere the neutron originated from, most of the time.
The data analysis is tricky, the neutron detection rates will be low, and if I was designing this I'd go for a thick secondary detector and count on thermalization and capture to create the secondary signal, rather than having a thin secondary detector looking for another recoil event. With a segmented detector or similar you'd be able to still do a reasonable job of the kinematics.
Discriminating against cosmic ray neutrons is going to be painful for this technology, however, and furthermore the comment that another poster made that "this tech shows we don't need to give up our civil liberties to be safe because it proves we can catch stuff at the boarder" is to my mind utterly wrong-headed. It assumes the border can be made perfectly impermeable, and that is simply not the case, as a million kilos of grass or whatever it is a year proves. As long as there is a chance that one bad guy can slip something through, Americans have two choices: be willing to die for your freedom, or give up your freedom (and be willing to die anyway, because a police state will not protect you.)
Final thought: we used to joke, back in the day, that we could sell our detector design to the U.S. navy as a means of detecting stationary nuclear submarines (it took a couple of days for useful neutrino statistics to build up when the prototype detector was about 10 m from a reactor core.) It looks like Nick might have found a way to do something very close to that after all...
Blasphemy is a human right. Blasphemophobia kills.
How long? let's see. if they're "developing" it now, say 3 years until it's in production and another year until it's at the major points of entry. But you've got to cover all the points of entry - sea, air(freight), land via Canada and Mexico. Make that about 10,000 PoE in all, so you're talking about another couple of years at least. That means about 6 years or until the beginning of 2014 to get a few grapefruit sized pieces of metal across the border.
Really bad thought: maybe it would be easier to get material that's already in US stockpiles - what use are border checks then?
politicians are like babies' nappies: they should both be changed regularly and for the same reasons
No, the range of these neutrons is pretty short. If unshielded and unscattered - a few hundreds of yards in free air. Or, in practice, a few tens of feet at best. I think you are confusing neutrons (which can be shielded against, but the shielding is heavy) with neutrinos (which go through everything).
Yes, this is true. This is not however an argument against these machines - as no defense will cover all bases.
Okay, let's see here...strokes kill about 150000 people each year in the U.S., and the government spends about $400 million on stroke research. Terrorists with radioactive materials have killed approximately zero people ever, and the government spends $650 million+ on (admittedly clever) directional radiation detectors. Yeah, we're clearly doing a great job of rationally allocating our money.
You fail to understand statistics. The 150k strokes a year is a large statistical sample and thus it is easy to predict the number from one year to the nect with some degree of statistical accuracy.
Now consider the nuclear case. There have been zero incidents since nuclear weapons existed in man-portable form, say 20 years ago. Now assuming a poisson distribution of events this means that we can conclude with a 95% confidence level that the rate of such events is less than 3.09 per 20 years i.e. less that a roughly 15.4% probability per year. Assuming that such an event would kill 1 million people this means that we are only 95% certain that the annual death rate from such terrorism is less than the death rate from strokes.
However the above is a conservative estimate because technology is making it easier to build nuclear weapons so whereas the above calculation assumed a constant probability distribution of such events that is not correct and it is getting more and more probable. So really we are less than 95% certain. In addition comparing the death rate is not a fair statistic. A better comparison would be years of human life lost. A majority, but certainly not all, stroke victims are old or have recently suffered other life threatening conditions like a heart attack or aneurism. However a terrorist bomb would kill children as much as the elderly (and everyone in between).
So while you cannot show that this is the most effective way of spending money to save life neither can you show that it is not. However given the uncertainties in any such calculation it is far from a total waste of time which is what you were suggesting.