Volume 103, Issue 2

Absorption-based X-ray micro-tomography (X-μCT) provides fundamental in-situ information on the 3D microstructure of complex multiphase materials such as cements. However, since the phases present in a hydrating cement paste may be characterized by similar values of the attenuation coefficient, leading to low absorption contrast between different crystalline or amorphous phases, micro-structural interpretation can be equivocal. 3D phase mapping by X-ray diffraction micro-tomography proved to be a successful technique for investigating the spatial distribution of the products in the paste during the hydration process, in a totally non-invasive mode and with enhanced phase selectivity compared to absorption tomography. Phase-selective maps, in the case of crystalline phases, can be extracted from single Bragg peaks or from the Rietveld-refined scale factor. However, even poorly crystalline and/or amorphous phases present in the cement paste, such as calcium silicate hydrates, can be successfully mapped by the use of selected portions of the measured powder data containing the relevant scattering of the phase. The reconstructed maps can be directly modeled by multifractal analysis and compared with computer-generated distributions.

Cold neutrons have a wavelength that is in the same range as the lattice spacings of most polycrystalline metallic materials. Imaging of such materials with monochromatic cold neutrons of different wavelengths provides a unique contrast due to coherent Bragg scattering. Additionally, the spectral positions of the Bragg edges can be mapped for each point of an image by using the transmission data of corresponding wavelength scans. We present investigations of welded components with such energy-selective neutron radiography around specific Bragg-edges in the transmission spectrum. Features in the local microstructure of the weld have been visualized.

We present methods for the geometric characterisation and stochastic modelling of open foams based on tomographic 3D image data. In the first step, geometric characteristics of the foam microstructure are estimated from the image. Using these characteristics, a random tessellation model is fitted to the cell system of the foam. The edges of this tessellation then serve as a skeleton for the foam's strut system. The focus of the paper is on the correct simulation of the foam's locally varying strut thickness in the model. For this purpose, the local strut thickness and the strut profiles are estimated from the image and reproduced in the model using locally adaptable morphology.

Optimising the manufacture of powder-processed components requires an understanding of the 3-D movement and behaviour of granular materials during processing. X-ray microtomography has been employed to study the 3-D flow behaviour of a metallic powder in-situ within a die as a function of the displacement of a punch into the die. In particular, the powder transfer behaviour for various open and closed die/punch geometries has been compared, including situations where features exist within the die and on the punch, cases where the die is both open and closed at the top, and finally where the punch itself contains a groove in the centre providing a gap into which powder can flow. Digital image correlation (DIC) has enabled the determination of local vector displacements of powder around the features within the die cavity as a function of punch movement and powder constraint to reveal bulk granular movement and densification. Zones of relatively stagnant flow are observed above a fixed insert within the die cavity, at the opening of a gap within a punch, and as a result of a closed die configuration, the latter showing transitions between the stagnant zones and much more mobile regions and the resulting powder compaction/dilation. As well as providing a means of developing practical die fill and compaction strategies that homogenise densification and thus improve product quality, the technique can also provide unique 3-D flow trajectories for model development.

Computed laminography has been introduced to synchrotron and neutron imaging set-ups to complement computed tomography for three-dimensional imaging of laterally extended (i. e. plate-like) specimens. The wide application field of computed laminography due to different contrast modes (X-ray or neutron absorption and X-ray phase contrast) and spatial resolutions ranging from some 100 down to approximately 0.5 μm is demonstrated. Selected examples from device inspection and from materials science are reported. They outline the interest of the method for non-destructive and in-situ measurements of regions of interest in large planar specimens where engineering-relevant boundary conditions have to be met. With a materials science background, the in-situ investigation of crack propagation in aluminium sheets and carbon-fibre composite panels under mechanical loading is reported.

A novel approach to strongly suppress artifacts in radiography and computed tomography caused by the effect of diffuse background signals (“backlighting”) of 2D X-ray detectors is suggested. Depending on the detector geometry the mechanism may be different, either based on the optical scattering by the fluorescent screen materials into optical detection devices or Compton or X-ray fluorescence scattering by the detector components. Consequently, these erroneous intensity portions result in locally different violations of Lambert–Beer's law in single projections (radiographs). When used as input data for computed tomography these violations are directly observed via modulation of the projected mass as a function of the rotation phase and the sample's aspect ratio (dynamics). The magnitude of the diffuse background signal depends on the detector area covered by the projected sample. They are more pronounced the smaller the shadowed area and the stronger the total attenuation. This implies that the reconstruction suffers from additional anisotropic artifacts caused by elongated sample structures. This issue is studied simply by absorption of flat plates in a conventional laboratory radiography set-up and at a synchrotron radiation facility. In the latter case beam hardening artifacts can be excluded due to the monochromatic radiation. The proposed correction procedure requires simple integral intensity offsets as a constant (non-local) light scattering mechanism is assumed.

The macroscopic properties of many materials are determined by structural features spanning a wide range of length scales. The nano-imaging endstation ID22NI of the European Synchrotron Radiation Facility is optimized for nanoscale imaging (X-ray focus size well below 100 nm) at high X-ray energy (17 – 29 keV), making it possible to study the interior of dense samples with a millimetric size. Computed tomography and computed laminography setups allow three dimensional imaging of a wide variety of samples, laminography being especially suited for non-destructive imaging of flat samples. The focused X-rays are used as a scanning probe for fluorescence tomography, providing sensitivity for elemental composition down to the ppm level. Projection microscopy with phase contrast is used for imaging the electron density in an extremely sensitive way. We demonstrate these capabilities by imaging a Be pebble. We found large scale cracks and nanoporosity (< 1 μm pores). Furthermore, beryllium oxide inclusions could be detected, ranging from 200 nm to several microns in size. Some of the smallest inclusions were located next to a small pore, indicating beryllium oxide formation after solidification of the structure.

Percolation is an important property of porous media, as it describes the connectivity of pores. We propose a novel, direction-dependent percolation probability which can be efficiently estimated from three-dimensional images obtained by microtomography. Furthermore, in order to describe the penetrability of the pore space by particles of a given diameter or a fluid of a given surface tension, we introduce a percolation probability depending on the width of the pores, from which we may also derive a measure of the mean pore channel width. As application examples, we consider the penetrability of porous beryllium pebbles, the connectivity of pores in arctic firn, the percolation of the pore space of aluminum foams and the mean width of the percolating space between the fibers in a laminate's percolating pore space.

For high performance, safety-critical applications, such as aerospace components, in-depth understanding of the material's response to complex loading conditions is essential. Moreover, it is vital to know how the material behaviour may be modified as a consequence of fatigue loading and how its eventual failure occurs. Unlike bulk properties, such as stiffness, yield stress, etc. that depend on the average response of the grains in a polycrystal, material failure is determined by “weakest link” type mechanisms. These depend strongly on grain-level deformation behaviour and grain-to-grain interactions. Micro-beam Laue diffraction is a powerful tool to probe these phenomena. However, the classical setup is limited to the study of sample surface regions or thin sections, due to the limited penetration into the sample at photon energies of 5 – 25 keV. A much more useful tool for the material scientist and engineer would allow the probing of grain-level orientation and stress in thicker sections of engineering components. To this end, we have developed the high energy transmission Laue (HETL) technique, an extension of the micro-beam Laue technique to significantly higher photon energies (50 – 150 keV). For the imaging of lattice orientation and elastic strain in three dimensions, we propose two alternative approaches: Laue orientation tomography (LOT) and high energy differential aperture X-ray microscopy (HEDAXM). In this paper an overview of the recent progress in HETL, LOT and HEDAXM measurements will be given and some first results illustrating the potential of these techniques presented.

Alpha-beta-tri-calcium phosphate is a promising biomaterial to fabricate scaffolds for the reconstruction of osseous defects. In the present study, we demonstrate that three-dimensional printing allows tailoring the porosity and the mechanical properties of nature-analogue designed scaffolds. Sophisticated software tools applied to the acquired tomography data were used to characterize the printing quality and the porosity by means of the distance transform, the radial density distribution and the accessibility by virtual spheres. Analyzing the stress – strain curves of the scaffolds, the Young's modulus and the ultimate compressive strength were obtained. To reach mechanical properties comparable to those of jaw bone, the scaffolds were treated with phosphoric acid. Depending on the design, the compressive strength of the bone scaffolds lies between 30 and 60 MPa and their Young's modulus is between 50 and 140 MPa. Hence, prototypes for clinical trials can be fabricated in near future.

The present study deals with in-situ microgap measurements in the internal taper connections of dental implants. Using X-ray phase contrast microtomography, the connecting interface between implant and abutment is probed non-destructively in three dimensions. Interference fringes across the conical interface occur due to the presence of microgaps, their intensity being a measure of the gap's width. Thus, for each point on the interface, interferences are extracted from the volumetric image in terms of normal projection maps which are, at selected points, compared to forward simulations to quantify the local gap width. Four designs of dental implants are tested in the “as-received” state as well as after application of cyclic extra-axial load. Results show different degrees of microgap opening by cyclic deformation according to the implants' design as well as a great amount of detail on the actual interface, i. e. fretting scars, grooves and wear debris.

Tomographic techniques are a valuable analytical tool as they deliver 3D spatial information on a given specimen. Both computed tomography with high spatial resolution and quantitative volume image analysis have made enormous progress during the last decade. In particular for materials and natural science applications the combination of high-resolution three-dimensional imaging and the subsequent image analysis exploiting the fully preserved spatial structural information yield new and exciting insights. In this paper, field-tested and up-to-date methods for tomographic imaging of microstructures, for processing and for quantitatively analysing three-dimensional images are reviewed. By selected applications from materials research, we shall underline the importance of volume image analysis as a crucial step in order to go beyond the images: it allows determination of spatial cross-correlations between different constituents of a specimen, investigation of orientations or derivation of statistically relevant information such as object size distributions. The core part of this work consists, besides the exemple application scenarios, in the processing chain, the tools and methods used.

Damage localization during power law creep of copper has been investigated in situ with synchrotron X-ray microtomography. The analysis of the area fraction of cavities corresponding to a given material slice has revealed that damage localization begins relatively early at half of the creep lifetime. The amplitude of the maximum fluctuation shows parabolic behavior as a function of mean void volume fraction. Existing models of damage evolution underpredict the amount of real damage and overpredict power-law creep life-time.

Predicting the fatigue lifetime of components relies on a knowledge of the residual elastic strain present throughout the bulk of the material. Non-destructively mapping the complete strain distribution throughout large volumes presents significant practical challenges. Recently a technique known as Bragg-edge neutron transmission has been developed as a means of non-destructive bulk elastic strain evaluation. Whilst conventional radiography measures the integral absorption, Bragg-edge neutron transmission probes the average strain along the incident beam direction. A “strain radiogram” is thus a two-dimensional average projection of the strain within the sample. Here we demonstrate how strain radiograms can be used for “neutron strain tomography” and we present and contrast two different approaches to the problem of characterising spatially resolved elastic strains.

The paper shows the benefit of registration tools in the quantitative analysis of tomography data. For this purpose three-dimensional datasets, i.e. target design in rapid prototyping, hard X-ray tomography, and magnetic resonance tomography, were registered with rigid, affine and non-rigid algorithms. Using rigid registration, we aligned individual tomograms to obtain one three-dimensional dataset per specimen, compared human teeth to determine abrasion, and optimized the acquisition parameters in clinical computed tomography. Affine registration was used to determine the sintering-induced shrinkage of hydroxyapatite scaffolds. Using non-rigid registration, we gained insight into tumor morphology imaged by different modes of micro computed tomography. In addition, the fixation-induced shrinkage of tumor tissue and the local deformation of brain tissues as well as the shape changes in nickel titanium alloys were quantitatively determined. Three-dimensional registration is, therefore, a powerful means to count back any influence on morphology of tissues and to quantitatively compare data of different tomography modalities.

The filtered medial line operator is a methodological tool for the morphological and topological analysis of reconstructed samples exhibiting either a porosity network or a granular structure such as bead packs. An application example of this methodology is the evaluation of the pore structure and the derived topological parameters of nanocrystalline 4Y-ZrO 2 specimens simulating the high burn-up structure of UO 2 fuel. The data indicate the absence of percolation paths at the experimental resolution, which is a beneficial property in terms of retaining fission gasses during the in-pile operation of the fuel. Furthermore, the analysis revealed a tendency for pore coalescence, which could explain the departure from the ideal behaviour of some physical properties typical of material matrices with a population of merely spherical pores. The same analytical tool is also used to characterise the porosity network created by gas bubbles in neutron irradiated beryllium samples, a material meant to be used for the blanket of the forthcoming generation of Tokamak fusion reactors. In particular, the preliminary results of the quantitative analysis performed on the porosity of the beryllium matrix as well as the initial investigation of its percolation properties are reported.