Latest developments and opportunities for 3D analysis of biological samples by confocal μ-XRF

https://doi.org/10.1016/j.radphyschem.2009.04.034Get rights and content

Abstract

X-ray fluorescence analysis performed with a primary radiation focused in the micrometer range is known as micro-X-ray fluorescence (μ-XRF). It is characterized by a penetration depth higher than other micro-analytical methods, reaching hundreds of micrometers in biological samples. This characteristic of the X-ray beam can be employed in 3D analysis. An innovative method to perform 3D analysis by μ-XRF is the so-called confocal setup.

The confocal setup consists of X-ray lenses in the excitation as well as in the detection channel. In this configuration, a micro-volume defined by the overlap of the foci of both X-ray lenses is analyzed. Scanning this micro-volume through the sample can be used to perform a study in three dimensions.

At present, X-ray lenses used in confocal μ-XRF experiments are mainly glass capillaries and polycapillaries. Glass capillaries are used in the excitation channel with sources of high photon flux like synchrotron radiation. Half polycapillaries or conical polycapillary concentrators are used almost exclusively in the detection channel. Spatial resolution of the confocal μ-XRF depends on the dimensions of the foci of both X-ray lenses. The overlap of these foci forms an ellipsoid which is the probing volume of the confocal setup. The axis length of the probing volume reported in confocal μ-XRF experiments are of order of few tens of micrometer.

In our confocal setup, we used a commercial glass monocapillary in the excitation channel and a monolithic half polycapillary in the detection channel. The polycapillary was home-made by means of drawing of multibundles of glass capillaries in a heating furnace. The experiment was carried out at the beamline D09B-XRF of the Synchrotron Light National Laboratory (Laboratório Nacional de Luz Síncrotron, LNLS) using white beam.

A model for the theoretical description of X-ray fluorescence intensity registered by confocal μ-XRF was introduced by Malzer and Kanngieβer [2005. A model for the confocal volume of 3D micro-X-ray fluorescence spectrometer. Spectrochim. Acta B 60, 1334–1341]. These authors showed that the measured scanning of fluorescent intensity is the spatial convolution of the sensitivity and the X-ray fluorescence emission rate detected. In a previous work we demonstrated that the convolution theorem can simplify the calibration and quantification process in confocal μ-XRF. In the present work, we applied these ideas to analyze samples of aquatic plants and a sample of tooth by confocal μ-XRF. We showed that confocal μ-XRF can be successfully applied to help the study of the effects of water bioremediation on aquatic plants. We also showed that confocal μ-XRF can be used to make topological studies of teeth.

Introduction

X-ray lenses have solved many problems in several X-ray analytical techniques. In XRF they have produced significant improvement in microanalysis by means of focusing lenses. X-ray lenses can focus photon beams in small spots increasing the intensity of the beams up to three orders of magnitude. Nowadays several devices to focus X-ray beams have been reported (Attwood, 1999). Among them, the most versatile are polycapillary lenses (Kumakhov and Komarov, 1990). They can be used to implement microanalysis by X-ray fluorescence (μ-XRF) in powerful sources such as synchrotron radiation as well as using conventional sources. Their applications appear particularly promising in case of techniques that make use of polychromatic X-ray beams. Indeed, they are able to transport X-rays of different energies in a broad energy bandpass. This property allows the expansion of the μ-XRF to three dimensions with the help of the so-called confocal setup.

In general, the confocal setup is developed to analyze a small volume of sample. The probing volume is defined by the overlapping of the foci of two lenses, one in the excitation channel and the other in the detection channel. This concept was successfully applied in conventional microscopy prior μ-XRF (Pawley, 1995). The scanning of the probing volume inside the sample allows the three dimensional analysis.

In μ-XRF the incident X-ray beam passes through a lens that focuses it into a small volume sample producing X-ray fluorescent emission. To focus the incident beam, a variety of X-ray optics has been employed. Monocapillaries (Balaic et al., 1996; Woll et al., 2006), polycapillaries (Janssens et al., 2004), curved graphite crystals (Bjeoumikhov et al., 1998) and compound refractive lenses (Vekemans et al., 2004) have been used with synchrotron radiation. In fact, any type of X-ray optic can be used with powerful sources of low divergence such as synchrotron radiation or free electron lasers. In contrast, for conventional sources with a high divergence (as X-ray tubes), the options are restricted to polycapillary optics (Kanngieβer et al., 2005; Xiaoya et al., 2008).

For the detection channel, the X-ray optics has to collect the emitted X-ray fluorescence from the probing volume with high divergence and wide energy range. Thus, the lens in the detection channel has to allow a wide acceptance angle and a broad energy bandpass. Polycapillary lenses are the ideal candidate to successfully satisfy these requirements. There are two types of polycapillaries, half lenses and conical collimators (Janssens et al., 2004; Bjeoumikhov et al., 1998), being the latter just the tip of a polycapillary lens.

As a consequence of the optics in the detection channel, the count rates detected are reduced. It is especially notable for background rates since only scattering photons coming from the probing volume are detected. Thus, XRF spectra in the confocal setup present a low background and the spectrum analysis is easier than in conventional XRF. Unfortunately, count rates of the characteristic radiation also decrease in such way that the lower limits of detection (LLD) are poorer than those of conventional XRF. For an optimized confocal setup, the LLD for metals can reach values in the range of 1 ppm to a few 10 ppm (Janssens et al., 2004). For a specific element, the LLD and the matrix absorption of its X-ray fluorescent emission determine the maximum depth achievable in the sample. For metals in biological samples this maximum depth is in the order of hundreds of micrometers.

Since up to now only polycapillary lenses can be used in the detection channel, they account for the minimum size of the probing volume in the confocal setup. At present, the minimum focus size of polycapillary lenses are in the submicron range, therefore the minimum axis length of the probing volume is in this range. When the axis length of the probing volume is greater than a few micrometer, the X-ray absorption inside the probing volume must be taken into account in quantitative analysis. For these cases, a theoretical formalism to obtain the primary X-ray fluorescent intensity was introduced by Malzer and Kanngieβer in (2005). These authors showed that the primary X-ray fluorescent emission in the confocal μ-XRF depends on the volumetric densities of the elements in the sample and not only on their weight fractions. Basically, this is due to the fact that the number of photons detected depends on the number of atoms inside the probing volume. The presented formalism can be used to develop new quantification strategies in confocal μ-XRF. As an example, a new formalism based on the convolution theorem has been developed recently (Pérez et al., 2008, Pérez et al., 2008).

Several applications of the confocal μ-XRF have showed the usefulness of the new analytical technique. The most frequent application has been the investigation of painted layers in art objects (Kanngieβer et al., 2003; Smit et al., 2004; Xiaoyan et al., 2008). Also the study of geological samples has been carried out (Vekemans et al., 2004; Janssens et al., 2004). In biology, the confocal μ-XRF has many potential applications (Bulska et al., 2006; Tsuji et al., 2007; Tianxi et al., 2008). For these types of samples, the absorption of X-rays in the matrix is small so that the theoretical description of the X-ray fluorescent emission can be simplified.

In the present work a description of the theoretical formalism required to study thin biological samples is shown. In this description, the mathematical expressions derived from the application of the convolution theorem are included. This facilitates the calibration process in the confocal μ-XRF. Finally the theoretical formalism is applied to study the rizofiltration with aquatic plants and the surface of a tooth.

Section snippets

Theoretical description of the confocal setup

In the confocal μ-XRF setup the X-ray fluorescence intensity of a specific X-ray line of an element i coming from the position x into an amorphous sample is given by:Isi(x)=0EmI0(E)τF,i(E)(Vρi(x)ηi(E,x-x)×exp[-Pathpμi,eff(p,E)dp]d3x)dEwhere Em is the maximum energy of the incoming photons, I0(E) the incoming photon flux of energy E, ρi the density (in g/cm3) of the i-element in the sample, τF,i the production cross section (in cm2/g) for the respective X-ray line of the i-element

Instrumentation

The experiment was carried out in the D09B beamline of the Synchrotron Light National Laboratory (Laboratório Nacional de Luz Síncrotron, LNLS) using white beam (Perez et al., 1999). A silicon drift X-ray detector with 150 eV of resolution at 5.9 keV was positioned perpendicularly to the photon beam on the horizontal plane. This system was mounted on a motorized XYZ stage. Suspended from the snout of the silicon drift detector, a fixed holder held a half monolithic polycapillary with its optical

Results and discussions

Fig. 2(a) and (b) show the experimental linear scans of the primary X-ray fluorescent intensity in standard pure foils of Cu and Zn respectively. The continuous curve represents the best fitting of Eq. (4) to the experimental points by the least square method. For σCu the mean value was (45±3) μm, and for σZn it was (48±3) μm. The parameter ki diminishes from the Cu to Zn in such a way that the integral in Eq. (3) also diminishes. It means that the global sensitivity of the spectrometer for Cu is

Conclusions

Since the first confocal μ-XRF setup was reported till present, this new technique was applied in various fields of research. The applications include investigations in archaeometry, geology, materials science and biology. Particularly in biology the confocal XRF has many potential applications as a consequence of its high achievable depth in biological samples. To these advantages, the valuable characteristics of conventional μ-XRF as the non-destructive character and simple sample preparation

Acknowledgments

This work was partially supported by the Brazilian Synchrotron Light Source (under proposal LNLS D09B-XRF-6512/07) and CONICET from Argentina. We are also grateful to the whole CEPROCOR and LNLS team for perfect running conditions.

References (27)

Cited by (32)

  • Quantitative analysis of the elemental composition of ion liquid with confocal X-ray fluorescence based on peak to background ratio

    2019, Radiation Physics and Chemistry
    Citation Excerpt :

    Since polycapillary X-ray optics were proposed in last century (Kumakhov and Komarov, 1990), they have been applied to XRF, and a new technology, called confocal XRF based on polycapillary X-ray optics, has been rapidly developed for the quantitative analysis of elemental composition (Sun et al., 2009). In recent years, confocal XRF has been widely used in various aspects, which include archaeology, painting, biology and environmental science (Choudhury et al., 2016; Prokeš et al., 2018; Roberto et al., 2010; Trojek and Trojková, 2015; Fittschen and Falkenberg, 2011). Due to its non-destructive character for elemental analysis, this technology has a great advantage in depth profiling of the elemental distribution inside samples, which could involve selective analysis at a special depth and region (Wrobel et al., 2014; Nakano and Tsuji, 2010), as well as being employed in three-dimensional (3D) elemental quantitative analysis (Mantouvalou et al., 2012).

  • Authentication of two samples of ancient Chinese coins with component element depth analysis by confocal 3D XRF

    2018, Nuclear Instruments and Methods in Physics Research, Section B: Beam Interactions with Materials and Atoms
  • Confocal X-ray fluorescence micro-spectroscopy experiment in tilted geometry

    2014, Spectrochimica Acta - Part B Atomic Spectroscopy
  • Direct deconvolution approach for depth profiling of element concentrations in multi-layered materials by confocal micro-beam X-ray fluorescence spectrometry

    2013, Talanta
    Citation Excerpt :

    The confocal micro-beam X-ray fluorescence (confocal μ-XRF) technique is an analytical tool that enables examination of spatial distributions of elements within a sample with a resolution ranging from several up to tens of micrometers. The method was proposed in 1993 by Gibson and Kumakov [1] and since then many authors proved its capability for analyzing samples of different origin, such as pigment layers in art objects or elemental distributions in biological and environmental samples [2–5]. The technique has been used with spectrometers operated either with synchrotron radiation or the radiation generated by X-ray tubes [6–8].

  • Focusing systems for the generation of X-ray micro beam: An overview

    2012, Spectrochimica Acta - Part B Atomic Spectroscopy
    Citation Excerpt :

    For example, to investigate biological samples a micrometer beam size is mandatory. Several studies on this matter are available and one of the latest articles is the one from Perez et al. [79]. The authors apply the model that was previously developed by Malzer et al. [26] as well as the calibration and quantification process to analyze samples of aquatic plants and a sample of a tooth by 3D μ-XRF.

View all citing articles on Scopus
View full text