Hacker News Reader

Comment by jandrewrogers | Hacker News Reader

I am not up-to-date on the bleeding edge but that explanation doesn’t seem correct? The use of x-rays in analytical chemistry is for elemental analysis, not molecular analysis. (There are uses for x-rays in crystallography that but that is unrelated to this application.)

At an elemental level, the materials of a suitcase are more or less identical to an explosive. You won’t easily be able to tell them apart with an x-ray. This is analogous to why x-ray assays of mining ores can’t tell you what the mineral is, only the elements that are in the minerals.

FWIW, I once went through an airport in my travels that took an infrared spectra of everyone’s water! They never said that, I recognized the equipment. I forget where, I was just impressed that the process was scientifically rigorous. That would immediately identify anything weird that was passed off as water.

View on HN

Here's an article that talks about Dual-energy CT [1]. And another one talking about material discrimination using DECT [2].

[1] https://en.wikipedia.org/wiki/Spectral_imaging_(radiography)

[2] https://pmc.ncbi.nlm.nih.gov/articles/PMC2719491/

Neither of those articles seem to support the idea that you can do molecular analysis with x-rays. They are all about elemental analysis, which is not useful for the purpose of detecting explosives.

Not sure if they use dual-energy x-ray as in [0], but you don't need to if you take x-ray shot from different angles. Modern 3D reconstruction algorithms you can detect shape and volume of an object and estimate the material density through its absorption rate. A 100ml liquid explosive in a container will be distinguishable from water (or pepsi) by material density, which can be estimate from volume and absorption rate.

https://en.wikipedia.org/wiki/Dual-energy_X-ray_absorptiomet...

See also beepblap's comments further below where they elaborate on this a bit (it's not just simple dual-energy xray apparently).

Hm, isn't it enough to just detect water and flag everything else as suspicious?

If your liquid is 80%+ water (that covers all juices and soft drinks), it is not going to be an explosive, too much thermal ballast.

> FWIW, I once went through an airport in my travels that took an infrared spectra of everyone’s water! They never said that, I recognized the equipment. I forget where, I was just impressed that the process was scientifically rigorous. That would immediately identify anything weird that was passed off as water.

Something like 10 years ago, I had my water checked in a specialised "bottle of water checker" equipment in Japan. I had to put my bottle there, it took a second and that was it. I have been wondering why this isn't more common ever since :-).

No idea if it was an "infrared spectra machine" of course.

Cynically, it's so they can sell you another bottle on the secure side. If they spend money to give themselves a working mechanism to distinguish water from not-water, they lose the ability to create retail demand.

I understand the idea, but it's not completely true: I empty my bottle before the security check, and fill it after in a fountain.

Then you have successfully circumnavigated a problem that more forgetful people will run into head-first. It doesn't have to catch everyone for the shops who are tenants on the secure side to complain about lost sales.

There's still no evidence that peroxide-based explosives are stable enough to be practical. And nobody every explained why the few liquid ones are so dangerous, but the solid ones get a pass when they are more stable.

It's a good thing that airport brought some machinery to apply the rule in a sane way. But it's still an insane rule, and if it wasn't the US insisting on it, the entire world would just laugh it off.

Reply on news.ycombinator.com