Colour X-ray machine sees so much more

Cnpower writes: "I dont understand this words, why use so much thickness filter? 20cm?"

-----

Hello cnpower,

It's not a filter in spite of what the article says. It's an x-ray collimator, i.e., it is a device designed to block all x-rays at the detector except those arriving from a particular direction. Tungsten is used for the collimator body because it is more opaque to x-rays than most other materials.

X-rays don't pay much heed to ordinary visible-light optics such as the kind of lens used in an ordinary optical collimator. As X-rays easily penetrate into or pass right through most materials, a different approach has to be used.

To get an idea of how an x-ray collimator works, hold a tightly-packed bundle of drinking straws at arm's length and look through one end of the bundle. So long as your line of sight is aligned with the long axis of the bundle you can see through it. But turn the bundle just slightly and your vision is blocked by the straws. The straws severely limit your field-of-view. Using longer and/or smaller-diameter straws restrict your field of view even more. Using shorter and/or larger-diameter straws, less.

X-ray collimators work exactly the same way except that role of the straws is played by an array of small holes bored through the length of the collimator body. X-ray collimators generally are not machined from a single piece of metal. Usually they're made from a stack of precisely-aligned metal plates in which identical sets of tiny holes have been drilled. The holes themselves are usually very, very small; on the order of tens of microns in diameter. Their small size places constraints on the maximum thickness of the material that can be machined, and usually this thickness is much less than the required length of the collimator. Consequently, I seriously doubt that the x-ray "filter" mentioned in this article was made from a single piece of metal. It's just not a practical approach. Try boring a single 50u-diameter hole (say) through 20 cm of solid tungsten. Then do it 255 more times and you'll see what I mean...

So that's pretty much what it is, what it does, and how it works. The "filter" has to be thick to limit the detector's field of view. Wanna guess why?

Best wishes,
-e

I knew I should call europium for this.

cnpower, please flog me, for my apparent lack of imagination...

Hi, Eman,

Good explain. originally I think that the filter would be acting as some meterial like Be which assemble in front of x-ray tube to reduce scatter radiation or some interference to sensor.

I know the collimator (aligner) was made of Lead and tungsten to limit the spread angle of x-ray to force to form into a line beam. but this seems to be a new situation

Im out of it. I shall learn slowly and if have any problem I shall ask yuou again.

Thanks.

Hello Cnpower,

I suspect that the detector itself discriminates between different x-ray photon energies. It's well known that higher-energy x-ray photons penetrate a given material to a greater depth than lower-energy photons, all else being equal, and this fact can be used as one way to discriminate between photon energies.

Semiconductor junctions can detect x-rays. Consequently one might build the detector as a stack of semiconductor junctions that measure how deeply an x-ray photon penetrates. The detector measures the penetration depth by noting the deepest junction that generates a signal. All junctions above that one may also generate a signal, depending on how the detector material responds to x-ray photons, but either way we know where the photon "landed." If the photons tend to ionize the material, there's a good chance that every junction down to penetration depth will fire. Junctions deeper than penetration depth will of course not produce a signal. A rectangular array of these stacks could be used to create a 2D "color" x-ray image.

Another approach might be to use arrays of atoms in a crystal lattice as a kind of x-ray diffraction grating to "sort" the incoming x-ray photons by energy. As the dispersion angle depends entirely on the energy, we can accurately differentiate the energies (energy ranges, actually) by placing simple detectors at a series of angles with respect to the grating. In other words, we build an x-ray spectrograph. To create a 2D x-ray image, we build an imaging spectrograph. A bit more complicated, but the idea's the same.

Cheers!

"...stack of semiconductor junctions that measure how deeply an x-ray photon penetrates..."

How can we be assured that the intermediate interaction half-way through the stack, wouldn't effect the scale of measurement, sort of quantum "measurement changes the measured object to warp the outcome".

After all, We are talking measurements here, linear or log scaled, but on the continuous un-interrupted column or scope.

OR

How is such detector or sensor, build in actual common practice, and how will it correlate wavelength (or energy level) to penetration, in a consistent way, un warped by actual particle to particle interaction, in the cascade described here?

How can we be assured that the intermediate interaction half-way through the stack, wouldn't effect the scale of measurement, sort of quantum "measurement changes the measured object to warp the outcome".

After all, We are talking measurements here, linear or log scaled, but on the continuous un-interrupted column or scope.

-----

The stack would have to be characterized - calibrated - empirically. The most straightforward way to do this would be to irradiate the stack with 'monochromatic' x-rays (ie, photons whose predetermined energies fall within a very narrow range), and correlate that range energy with the measured penetration depth. Then change the energy range and measure the penetration depth at that range, and so forth, until the stack is completely characterized.

In a sort of "Trial And Error" re-iteration sequence?

In a sort of "Trial And Error" re-iteration sequence?

-----

Yes, but no self-respecting scientist would call it that. "Trial And Error re-iteration sequence" sounds too much like a procedural crapshoot, so they've come up with a glorified euphemism for it that means the same thing: "Research and Development."

Hi,Eman,

your idea is right. why dont you continue this research to complete the product, I think there will be wild use in practice.

there are many peoples dedicating on the subject now,