Quantitative comparison of X-ray fluorescence microtomography setups: Standard and confocal collimator apparatus ☆ M. Chukalina a , A. Simionovici b, ⁎ , S. Zaitsev a , C.J. Vanegas a a Institute of Microelectronics Technology RAS, 142432, Chernogolovka, Moscow District, Russia b Laboratoire de Géophysique Interne et Tectonophysique, University of Grenoble, BP 53, 38041, Grenoble, France Received 3 March 2006; accepted 18 February 2007 Available online 2 March 2007 Abstract Recently, there has been a renewed interest for fluorescence spectroscopy, as provided by modern setups which allow 2D and 3D imaging of elemental distributions. Two directions are currently under development: the SR-based fluorescence tomography in polar scanning geometry, provided by the new generation of X-ray microprobes and the confocal scanning geometry, which can be fielded in both SR and laboratory environments. The new probes bring forth a new age in fluorescence spectrometry: high resolution, high intensity and high sensitivity which allow 3D elemental mapping of volumes. The major task now is the development of these complex tools into fully quantitative probes, reproducible and straightforward for general use. In this work we analyze two X-ray fluorescence microtomography techniques: an apparatus tomography using a confocal collimator for the data collection and a standard first generation Computed Tomography (CT) in the parallel scanning scheme. We calculate the deposited dose (amount of energy deposited and distributed in the sample during the data collection time) and find the conditions for the choice of the tomography scheme. © 2007 Elsevier B.V. All rights reserved. Keywords: X-ray fluorescence computer tomography; Deposited dose; Confocal scanning 1. Introduction X-ray fluorescence micro-computed tomography is one of the modern techniques, which allows investigating chemical elements distribution inside a sample without any destructive or invasive sample preparation. The main concept of the X-ray fluorescence technique is the following: X-ray radiation illuminates a sample and a characteristic element X-ray line is measured. To obtain a local information about the element distribution inside the sample, different tomography scanning schemes for the collection of tomographic projections are used, then from the set of projections the element distribution is reconstructed by a special mathematical procedure. In this work we present a comparison analysis of two different approaches. The first one is an apparatus X-ray fluorescence microtomo- graphy as shown in Fig. 1. This device is similar to the polycapillary half-lens setup recently developed and fielded by several groups [1–7]. The X-ray microbeam illuminates part of the sample, which generates the fluorescence. Using a confocal collimator with “N” channels, placed between the sample and the detector, a unit volume (voxel) in the sample is localized at the intersection with the focused incident beam, so-called vertex voxel. Focusing of the primary beam is needed only to decrease the data collection time and the background produced by scattering in the whole sample, but it is not a mandatory condition of the experiment. It has been shown [8] that the main contribution to the recorded signal is given by the vertex voxel. In this way it is possible to obtain information about the element under investigation voxel by voxel alternatively moving the sample in front of the X-ray beam, thus scanning the vertex Spectrochimica Acta Part B 62 (2007) 544 – 548 www.elsevier.com/locate/sab ☆ This paper was presented at the “18th International Congress on X-ray Optics and Microanalysis” (ICXOM-18) held in Frascati, Rome (Italy), 25–30 September 2005, and is published in the Special Issue of Spectrochimica Acta Part B, dedicated to that conference. ⁎ Corresponding author. E-mail addresses: marina@ipmt-hpm.ac.ru (M. Chukalina), alexandre.simionovici@ujf-grenoble.fr (A. Simionovici), zaitsev@ipmt-hpm.ac.ru (S. Zaitsev), vanegas@ipmt-hpm.ac.ru (C.J. Vanegas). 0584-8547/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.sab.2007.02.020