Spectral-sensitive proton radiography of thin samples with the pixel detector Timepix3

Abstract

We evaluate a high-resolution contrast-enhanced method for energy-sensitive radiography of thin samples with low-energy protons at the light-ion Tandetron accelerator of the NPI-CAS in Rez near Prague. We make use of the high-sensitivity of the hybrid semiconductor pixel detectors Timepix enabled by integrated per-pixel signal processing electronics. For this work we use the Timepix3 ASIC chip equipped with a 500 µm Si sensor operated with the fast data rate AdvaPix readout electronics interface. Measurements are performed in air with a 2.9 MeV proton microbeam on thin samples (<100 µm thick). As referential and testing sample we use a set of aluminum foils stacked into a closely packed assembly of varying layers of well-defined thickness. This and other samples were imaged and placed in front of the detector in transmission geometry. Radiographies were collected with focused beam (few mm size) and a microbeam (few µm size). The imaging principle is based on high-resolution spectrometry of single transmitted particles. Contrast is obtained by registration of small differences in the deposited energy of the proton after passing through the sample. This can be measured in wide-range by detailed spectral-tracking analysis of the pixelated clusters in the pixel detector. We examine and evaluate various cluster-track parameters sensitive for imaging contrast such as deposited energy, cluster area (number of pixels) and cluster height (maximum energy value of the pixels in the cluster). The position of interaction in the detector is registered in sub-pixel resolution down to few µm scale for the particles and geometry used. Radiographies are reconstructed based on these individual parameters imaged in image bins of adjustable size (few µm up to few tens of µm). The technique developed with different cluster parameters is presented together with evaluation of image contrast sensitivity on various types of samples and beam energies.