Search

This research is focused on studying the tension-tension fatigue behaviour of a unidirectional (UD) glass-fibre wind turbine composite. The damage features, their progression and the associated strain fields are tracked in a representative volume by employing a novel correlative approach bringing together x-ray computed tomography (XCT) and digital image correlation (DIC). The focus is on studying ex situ the evolution of damage features (fibre breaks and micro cracks) in an interrupted time-lapse manner. The major drops in stiffness are correlated to the number and location of the damage features in the bulk (XCT) and at the surface (DIC). Results from XCT highlight a localized cluster of fibre breaks and matrix cracks near backing bundles along with axial macro-cracks, while DIC shows that the backing bundles cause regions of higher strain. This highlights the relation between the damage features and strain localisation and their effect on the progressive degradation in stiffness during high cycle fatigue (HCF) cycling.

Biomaterial for tissue engineering is a topic of huge progress with a recent surge in fabrication and characterization advances. Biomaterials for tissue engineering applications or as scaffolds depend on various parameters such as fabrication technology, porosity, pore size, mechanical strength, and surface available for cell attachment. To serve the function of the scaffold, the porous biomaterial should have enough mechanical strength to aid in tissue engineering. With a new manufacturing technology, we have obtained high strength materials by optimizing a few processing parameters such as pressure, temperature, and dwell time, yielding the monolith with porosity in the range of 80%&ndash, 93%. The three-dimensional interconnectivity of the porous media through scales for the newly manufactured biomaterial has been investigated using newly developed 3D correlative and multi-modal imaging techniques. Multiscale X-ray tomography, FIB-SEM Slice &amp, View stacking, and high-resolution STEM-EDS electronic tomography observations have been combined allowing quantification of morphological and geometrical spatial distributions of the multiscale porous network through length scales spanning from tens of microns to less than a nanometer. The spatial distribution of the wall thickness has also been investigated and its possible relationship with pore connectivity and size distribution has been studied.

In the present study, time-evolved (4D) correlative 3D imaging techniques, combining microtomography and nanotomography with 3D FIB-SEM sectioning and EDS, have allowed shedding light on morphological and compositional aspects of the corrosion behaviour of AZ31 magnesium alloy. By applying these techniques, the corrosion processes involved, corrosion stages I, II and III, were identified successfully despite the challenging fast corrosion behaviour of magnesium-based alloys. The 3D reconstruction confirmed the near-surface restriction of the filiform corrosion (stage II) and the subsequent severe inward corrosion during stage III. Furthermore, the distribution of the coarse Al–Mn intermetallic particles could be correlated with the corrosion front development.

Lithium-ion batteries (LIB) are today the main energy storage devices from portable electronics devices to electric vehicles. LIB have a high energy density, low self-discharge, and low memory effect allowing for large number of charging cycle without degrading storage capacity. Performance, cost and safety are the main factors driving research and different chemistry are investigated in order to achieve better performance at a lower cost and increase safety. Research, development and quality control of LIB can benefit much from X-ray Computed Tomography (CT) and Focused-Ion Beam – Scanning Electron Microscopy (FIB-SEM). MicroCT allows for complete non-destructive acquisition of a complete battery cell, allowing to control the production process and to monitor the changes occurring in the structure during multiple charging cycles. Different visualization techniques are proposed on a 7µm microCT acquisition of a Lithium Nickel Manganese Cobalt Oxide cell 18650 (NMC) which allow for a visual inspection of the internal structure of the cell, including techniques of unfolding to look at rolled electrodes and measurement such as electrode thickness, height, or length. Macroscopic properties of a battery, including mechanical, thermal and electrical, find their origins at a nano-scale. Using FIB-SEM, high resolution images can be acquired which can reveal the structure of the electrodes or the separator. Besides the volume fraction or surface areas of the different phases, the connectivity of pores and/or particles, their surface of contacts, path tortuosity or constrictivity, permeability or molecular diffusivity, become accessible. We will present simulation results on permeability and tortuosity on this sample. We will also present correlative experiment, performing first a microCT scan of a cathode foil sample on its aluminium collector, spotting sites of interest in it, and acquiring those volumes with FIB-SEM.

Syntactic foams are widely used in many impact-absorbing applications and can be employed as sandwich core. To improve their mechanical performances, these composite sandwich structures have to be modelled. This approach requires the characterisation of the foam behaviour. Moreover, the microstructure of the syntactic foam has an influence on its macroscopic behaviour; the foam density, the diameter of the porosities, their distribution in the material have to be taken into account.

In this investigation, a computational model has been developed for simulation of low melting point alloys solidification process. The proposed model is capable of considering the effects of resistance of metal–mould interface and metalo-static pressure during solidification, jointly. The simulation of interface resistance has been achieved on the Zero Thickness Element (ZTE), utilizing the Finite Element Method. The solid boundary conditions, including contact resistances, have been modified by pressure gradient in each ZTE. The effect of pressure gradient has been modeled based on the experimental data, and an innovative model was developed. In order to verify the computational results, an Al–Si 11.5% alloy was poured into a permanent mould and the temperature of the interface was measured by data acquisition system for alternative metalo-static pressures. Then, the effect of metalo-static pressure was modeled as in the case of an overall heat transfer coefficient at the metal–mould interface. A comparison between the experimental and the simulation results during solidification process showed a good consistency, which confirms the accuracy of the present model for estimating the solidification time.

A numerical model has been developed for the determination of liquid flow permeability through columnar dendrite during growth and segregation in the Al–Si alloys. Therefore, in the present work two separate computational models of the grain growth and interdendritic liquid flow are coupled for modelling of the permeability in partly solid alloy. The grain growth is simulated using a cellular automation finite difference (CAFD) for a 2D dendrite, and fluid flow by using a computational fluid dynamic (CFD) model for determining permeability. A new model has been presented for calculation and modification of the dendrites permeability in the high solid fractions. Simulation results show which Si concentration variations in each time step could transform of dendrite shape. Also dendrites morphology could alter the interdendritic permeability factor.

The clustering of reinforcement particles in cast metal matrix composite (MMC) material has a number of deleterious effects on its mechanical properties and castability. In order to better understand this phenomenon, the clustering behaviour of TiB2 particles in molten commercial purity aluminium was investigated. A novel method of sampling the melt was developed in order to investigate the clustering behaviour as a function of holding time at 700 °C. The initially clustered distribution of TiB2 became increasingly dispersed with increased holding time. The maximum observed cluster size decreased from 50 to 10 μm after a holding period of 73 h. X-ray microtomography (XMT) and SEM micrograph analyses were carried out to quantify the rate of dispersion of the clusters. XMT was shown to be a powerful tool for visualizing and quantifying 3-dimensional features within the MMC microstructure.

Phase transition phenomena in metallic alloys involve complex physical processes occurring over a wide range of temporal, spatial and energy scales. Multiscale modelling is a powerful methodology for understanding these complex systems. In this paper, a multiscale model of grain and pore formation is presented during solidification. At the microscale, a combined stochastic-deterministic approach based on the cellular automata method is used to solve multicomponent diffusion in a three-phase system (liquid, solid and gas), simulating the nucleation and growth of both grains and pores. The impingement of the growing pores upon the developing solid is also solved to predict the tortuous shape of the porosity, a critical factor for fatigue properties. The micromodel is coupled with a finite element method (FEM) solution of the macroscale heat transfer and fluid flow in industrial castings through the temperature and pressure fields. The result model was used to investigate the influence of local solidification time, hydrogen content, local metallostatic pressure and alloy composition upon the predicted grain structure and pore morphology. Comparison of the model predictions to both laboratory and industrial scale castings are presented.

Multi-scale modelling of phenomena in which a complete description requires the coupling of many processes at widely different length, time and energy scales offers a new and state-of-the-art strategy in computational condensed matter physics. This review is concerned with a multi-scale computational modelling of the solidification phase transition, as a highly complex and truly multi-scale process. The modelling integrates the initial nucleation processes unfolding at the nano-scale with a probabilistic micro-scale model of micro-structure formation. The resulting nano–micro model is then coupled with the macroscopic heat flow equation to provide an unified approach to the solidification phenomenon. Molecular dynamics (MD) simulation method provides the theoretical framework for modelling the formation of the initial atomic clusters at the nano-scale. Phase transitions are detected at this scale, and the all important material properties are also computed. Pertinent inter-atomic potentials that model the energetics and dynamics of the clusterisation process at the nano-scale are also given. The computed material properties form the input into a cellular automata (CA)-based model of the micro-structure formation at the micro-scale, as well as the input to the macroscopic heat flow equation. Solid fractions generated from the micro-scale model also form part of the input to the macro-scale model whose temperature field distribution is, in turn, fed back into both the micro-scale and the nano-scale models. There is thus a close interplay between the different levels of the multi-scale model. As well as providing a nano-scale basis for the CA-based model of micro-structure formation, an extension of that model, based on concepts from the Ito stochastic dynamics, is also discussed. A set of computer-based simulations of the solidification of the elemental and alloy systems are discussed for each component of the overall multi-scale model, with the full multi-scale model applied to the simulation of the solidification of some of the industrially important alloy systems.

International audience; We produce cellular material based on the formulation of model emulsions whose drop size and composition may be continuously tuned. The obtained solid foams are characterized by narrow cell and pore size distributions in direct relation with the emulsion structure. The mechanical properties are examined, by varying independently the cell size and the foam density, and compared to theoretical predictions. Surprisingly, at constant density, Young's modulus depends on the cell size. We believe that this observation results from the heterogeneous nature of the solid material constituting the cell walls and propose a mean-field approach that allows describing the experimental data. We discuss the possible origin of the heterogeneity and suggest that the presence of an excess of surfactant close to the interface results in a softer polymer layer near the surface and a harder layer in the bulk.

12 pages; International audience; The mechanical properties of cellular materials are still subject to numerous theoretical and experimental investigations. In particular, the impact of cell size on the foam's elastic response has not been studied systematically mainly due to the lack of experimental techniques with which the cell size and relative density of materials can be varied independently. This paper presents the results of a study of the elastic behavior of open-cell foams as a function of relative density and the size of the interconnected, spherical pores. First, the chemical procedure allowed us to produce polystyrene open-cell foams in which the relative density and the average cell diameters were varied independently. The results of compression tests performed on these foams showed an unexpected influence of the cell diameter (at constant relative density) on the elastic response. The analysis of the microstructure of the foam revealed the presence of a complex nanostructure in the edge of the cells that appeared during the synthesis procedure. An analytical model (an extension of the Gibson-Ashby model) is presented, which takes into account the complex multi-scale structure of the foam and accurately describes the observed dependence of the measured Young's moduli on cell size. This approach was confirmed further by a finite element numerical simulation. We concluded that the observed dependence of elastic modulus on cell size was due to the heterogeneous nature of the material that constitutes the walls of the cells.

Using micro tomography, one of the most detrimental imaging noises to porous media characterisation is the ring artefact, which is mainly due to the detector's non-linearity and defects. Ring artefacts change artificially the porous phase microstructure. Therefore, ring artefacts' filtering is essential to any evolutional study of the porous phase, under a set of given stresses within a material system. In this paper, a new filtering method for ring artefacts is presented. The method is applied to the percolation study of a calcite-based porous rock. The choice of the material is motivated by the considerable increase of carbon dioxide emissions into the atmosphere and its environmental and economical implications, caused by the greenhouse effect. This, encouraged research on the transformation of geological spaces into carbon dioxide sequestration reservoirs.