The spectrum of a beam of therapeutic x-rays (megavolt) or diagnostic x-rays (kilovolt) cannot be measured precisely. And it seems that an estimation of spectra using deconvolution of attenuation data is the most common method to get such spectra. The method of estimating x-ray spectrum using measured attenuation data is an ill-posed problem on itself (a situation where we solve for larger number of unknowns with fewer number of equations). Using an estimated or derived spectra we can calculate back the attenuation data. A comparison of the calculated attenuation data with the measured attenuation data is the only method that I use and know so far.
Besides this, could anyone please suggest me any other idea to check the accuracy of a spectrum?
Thank you in advance.
Assume you have a standard detector, so it measures overall intensity of X-rays reaching it. Then you only have the material between the source and the detector to change.
You could use a range of pure element with particular K-edges, as it would absorb stronger for a particular energy range. Also you could vary your material thickness. You need a good data base also, "NIST: X-Ray Mass Attenuation Coefficients" is a good place to find that info.
You could solve for bands of X-rays, so how many in each 10 or 100 kilovolt range for example, it would make your calculation a bit easier.
Dear Rajesh Panthi
You can get the HVL of your beam in a high purity Al or Cu according to the kV of you x-ray beam.
From the HVL, calculate the corresponding attenuation coefficient and so the effective x-ray energy can be calculated by the aid of winxcom of NIST as Heyang explain.
Hi Rajesh Panthi,
many years ago I heard about an experiment in which direct the spectrum of a Tungsten x-ray tube was recorded using an energy dispersive detector... the idea was the attenuation of the intensity in a long tube (15 m??) containing several pin-holes (e.g. as used as apertures in TEM or SEM. I never try it.
Regards Alex
Dear Rajesh
The use of an EDS, at least for the softest part of the hard x-ray spectrum, is a valuable method as mentioned by Alex in combination with a set of filters both to calibrate the energy scale (energies of K edges are well-known) and to perform absolute intensity measurements : it is possible to find screens the thickness of which is accurately estimated. Comparaison with theoretical model for Bremsstrahlung (see for instance J Phys B 32 4261 1999) makes it possible to refine the measurements.
Best regards
As an additional comment the various responses, I should point out that the purpose of the question is maybe not ambiguous but certainly not focused on a particular application, and not quantitative: what you want to do depends on what you want out of if, and with what precision.
It is one thing if a spectral measurement were an exercise in the measurement technique, e.g., to confirm the well-known theory of bremsstrahlung production that is by now captured in Monte Carlo codes such as NRC-EGS (for the medical community mostly, I think), GEANT4 (for many physicists), ITS (for some in the defense community), MCNP, and others. Those have been extensively verified so it may be best to get the spectrum you need from such code.
The next issue is: do you really need the spectrum? or are you interested mostly in the dose resulting from a spectrum, and what is the necessary accuracy? The dose depends little on the shape of the spectrum but is proportional to the intensity, so that a good intensity measurement combined with a ration of absorption cross sections may be all you need.
On the measurement techniques: the best energy resolution you get with Bragg diffraction from crystals, which is how photon-counting energy dispersive detectors such as High-Purity Germanium detectors are ultimately calibrated.
- Badges
- Science topic
- Similar topics
- Correction