[lloraine (OP)]

Hi!

I've been doing some research on X-ray physics and I'm not sure about one thing - the application of both of spectra. One article said characteristic spectrum is used in mammography, the other said the opposite. While continuous spectrum is for basic X-rays. Maybe you guys know the truth? Source is more than welcome.

Kamila

[alancalverd]

We don't use the entire continuous spectrum for radiography.  Low energy photons are absorbed in superficial tissue so contribute to skin burning but not to image formation. High energy photons penetrate further through tissue but discriminate less between low atomic number (low Z) and high Z absorbers, so you get less contrast between bone (containing calcium and phosphorus) and soft tissue (mostly carbon and water). So we limit the x-ray tube voltage to less than 150 kV and remove the low energy part by filtration through aluminium or copper, leaving a spectrum with a peak intensity around 40 - 70 keV which forms the most useful images for general radiology. A tungsten anode yields a fairly clean initial continuous spectrum (the characteristic lines at 60 keV are a small fraction of the useful energy) that can be tailored to the  thickness of the part to be x-rayed by simply altering the tube potential.

In mammography we are trying to image subtle differences in soft tissue density and pick out tiny calcifications. This is best done with a very narrow spectrum to minimise surface dose (from a low energy component) and low-contrast fog (from high energy photons).  It turns out that the molybdenum Kα characteristic line at 17.5 keV does the job very well. If we restrict the anode voltage to less than 30 kV and absorb the higher energy part of the continuum with a Mo filter, we get a beam that is close to ideal as long as we can compress the breast to minimise the overall absorption and scatter of the useful photons.

It would be really great to get rid of the low energy component altogether, but whilst a radionuclide source would give us a genuine line spectrum, the effective source diameter would be too large -  we need a point source with a high dose rate to get a sharp image.