Three-dimensional Reconstruction Of Microstructures Research Articles

A new approach to three-dimensional microstructure reconstruction of a polycrystalline solar cell using high-efficiency Cu(In,Ga)Se2

A new method for efficiently converting electron backscatter diffraction data obtained using serial sectioning by focused ion beam of a polycrystalline thin film into a computational, three-dimensional (3D) structure is presented. The reported data processing method results in a more accurate representation of the grain surfaces, reduced computer memory usage, and improved processing speed compared to traditional voxel methods. The grain structure of a polycrystalline absorption layer from a high-efficiency Cu(In,Ga)Se2 solar cell (19.5%) is reconstructed in 3D and the grain size and surface distribution is investigated. The grain size distribution is found to be best fitted by a log-normal distribution. We further find that the grain size is determined by the [Ga]/([Ga] + [In]) ratio in vertical direction, which was measured by glow discharge optical emission spectroscopy. Finally, the 3D model derived from the structural information is applied in optoelectronic simulations, revealing insights into the effects of grain boundary recombination on the open-circuit voltage of the solar cell. An accurate 3D structure like the one obtained with our method is a prerequisite for a detailed understanding of mechanical properties and for advanced optical and electronic simulations of polycrystalline thin films.

Three-dimensional microstructure reconstruction for two-phase materials from three orthogonal surface maps

Introduction: A full three-dimensional (3D) microstructure characterization that captures the essential features of a given material is oftentimes desirable for determining critical mechanisms of deformation and failure and for conducting computational modeling to predict the material’s behavior under complex thermo-mechanical loading conditions. However, acquiring 3D microstructure representations is costly and time-consuming, whereas 2D surface maps taken from orthogonal perspectives can be readily produced by standard microscopic procedures. We present a robust and comprehensive approach for such 3D microstructure reconstructions based on three electron backscatter diffraction (EBSD) maps from orthogonal surfaces of two-phase materials.Methods: It is demonstrated that processing surface maps by spatial correlation functions combined with principal component analysis (PCA) results in a small set of unique descriptors that serve as a representative fingerprint of the 2D maps. In this way, the differences between surface maps of the real microstructure and virtual surface maps of a reconstructed 3D microstructure can be quantified and iteratively minimized by optimizing the 3D reconstruction.Results: To demonstrate the applicability of the method, the microstructure of a metastable austenitic steel in the two-phase region, where austenite and deformation-induced martensite coexist at room temperature, was characterized and reconstructed. After convergence, the synthetic 3D microstructure accurately describes the experimental system in terms of physical parameters such as volume fractions and phase shapes.Discussion: The resulting 3D microstructures represent the real microstructure in terms of their characteristic features such that multiple realizations of statistically equivalent microstructures can be generated easily. Thus, the presented approach ensures that the 3D reconstructed sample and the associated 2D surface maps are statistically equivalent.

Descriptor-based reconstruction of three-dimensional microstructures through gradient-based optimization

Microstructure reconstruction is an important cornerstone to the inverse materials design concept. In this work, a general algorithm is developed to reconstruct a three-dimensional microstructure from given descriptors. Based on two-dimensional (2D) micrographs, this reconstruction algorithm allows valuable insight through spatial visualization of the microstructure and in silico studies of structure-property linkages. The formulation ensures computational efficiency by casting microstructure reconstruction as a gradient-based optimization problem. Herein, the descriptors can be chosen freely, such as spatial correlations or Gram matrices, as long as they are differentiable with respect to the microstructure. Because real microstructure samples are commonly available as 2D microscopy images only, the desired descriptors for the reconstruction process are prescribed on orthogonal 2D slices. This adds a source of noise, which is handled in a new, superior and interpretable manner. The efficiency and applicability of this formulation is demonstrated by various numerical experiments.

Unsupervised Generative Adversarial Network for 3-D Microstructure Synthesis from 2-D Image

Three-dimensional microstructure reconstruction is indispensable for understanding the performance and reliability of SOFC electrodes. Not only the microstructure parameters, but also SOFC electrochemical characteristics such as overpotentials can be calculated when combined with numerical simulations. The most common method to analyze 3-D structures with a feature size of 0.01 – 10 μm is a focused ion beam - scanning electron microscope (FIB-SEM). However, there’s a tradeoff between reconstruction volume size and resolution. Besides, in order to obtain high quality reconstruction, accumulation of know-hows for pre and post-processing is required. Therefore, fast screening of the microstructures is not easy for FIB-SEM due to the difficulty of measurement. On the other hand, a single SEM image can be obtained very easily. An arising question is if it is possible to extract properties of microstructures and numerically synthesize 3-D model directly from a single 2-D SEM image.In recent years, machine learning techniques have become a powerful tool for image processing. Nevertheless, applications of machine learning in the field of porous materials, in particular fuel cells and batteries are still limited. The previous works concentrated on relatively simple artificial neural networks (ANN), but the class of neural networks that can effectively analyze visual images is the convolutional neural network (CNN). Gayon-Lombardo et al. (1) demonstrated a successful application of generative adversarial network (GAN) based on CNN layers to fabricate synthetic 3-D SOFC microstructures. It is indicated that the GAN-based approach can be very effective, but the algorithm requires 3-D teaching data (GAN3Dto3D).In the scope of this research, a novel method for fabrication of synthetic 3-D microstructure directly from a 2-D SEM image is proposed (Fig. 1A). The GAN2Dto3D network consists of a 3-D microstructure generator and a discriminator which operate on 2-D data. Each microstructure fabricated by the generator is treated as a set of individual cross-sectional 2-D images. The discriminator checks the validity of each 2-D slice in the generated microstructure in all spatial directions (height, width and thickness). Patches extracted from a real 2-D cross-section image are used as training data. In each training iteration, new random patches are extracted to increase the number of available training data. Additionally, the training images are augmented by random horizontal and vertical flipping. As all of the cross-section of the newly generated microstructure has to be statistically equivalent to the real 2-D cross-section, the generated 3-D microstructure is smooth in all spatial directions.The GAN2Dto3D approach is tested on the dataset of real SOFC microstructures. For the real microstructure, a set of nickel (Ni) – gadolinium doped ceria (GDC) microstructures was fabricated. All composite anode samples were prepared with NiO (AGC Seimi, Japan) and GDC (Gd0.1Ce0.9O ShinEtsu, Japan) powders mixed in a designated proportion and sintered at 1350 oC. Differences in the structure are due to the initial anode slurry composition and packing density. The solid phase fractions were controlled by GDC share in the composite in the range of 30 – 70 vol.%, and their porosity was controlled by isostatic pressing (2). 3-D microstructures were reconstructed by FIB-SEM and used as the ground truth data for the validation of GAN2Dto3D model. The results of the GAN2Dto3D algorithm applied to generate two synthetic microstructures with different GDC share and the porosity are shown in Fig. 1B and 1C. The size of the generated microstructures is 64 × 64 × 64 voxels, which is equivalent to 6.4 × 6.4 × 6.4 µm. The real microstructures reconstructed with FIB-SEM are shown for comparison. The proposed GAN algorithm produces realistic microstructures, which indicates that GAN has a great potential to enable fast evaluation of various samples without carrying out complex FIB-SEM measurements.1. Gayon-Lombardo, L. Mosser, N. P. Brandon, and S. J. Cooper, npj Comput. Mater., 6, 1–11 (2020).2. Komatsu, A. Sciazko, and N. Shikazono, J. Power Sources, 485, 229317 (2021). Figure 1

A Three-Dimensional Microstructure Reconstruction Framework for Permeable Pavement Analysis Based on 3D-IWGAN with Enhanced Gradient Penalty.

Owing to the increasing use of permeable pavement, there is a growing need for studies that can improve its design and durability. One of the most important factors that can reduce the functionality of permeable pavement is the clogging issue. Field experiments for investigating the clogging potential are relatively expensive owing to the high-cost testing equipment and materials. Moreover, a lot of time is required for conducting real physical experiments to obtain physical properties for permeable pavement. In this paper, to overcome these limitations, we propose a three-dimensional microstructure reconstruction framework based on 3D-IDWGAN with an enhanced gradient penalty, which is an image-based computational system for clogging analysis in permeable pavement. Our proposed system first takes a two-dimensional image as an input and extracts latent features from the 2D image. Then, it generates a 3D microstructure image through the generative adversarial network part of our model with the enhanced gradient penalty. For checking the effectiveness of our system, we utilize the reconstructed 3D image combined with the numerical method for pavement microstructure analysis. Our results show improvements in the three-dimensional image generation of the microstructure, compared with other generative adversarial network methods, and the values of physical properties extracted from our model are similar to those obtained via real pavement samples.

Reconstruction of Three-Dimensional Microstructures of Two-Phase Membrane and Phase Property Estimation Through Combination of Experiment and Simulation

In this study, a multiscale property evaluation framework for heterogeneous materials with complex microstructures on a sub-micro scale has been proposed. A detailed investigation of the three-dimensional (3D) microstructural features is necessary for the evaluation of the properties of heterogeneous materials with complex microstructures. However, for heterogeneous materials with microstructures on a sub-micro scale, it is difficult to observe the 3D microstructural features owing to the limit of resolution. In this study, virtual 3D samples were reconstructed based on two-dimensional images, and tensile simulations were performed using the reconstructed 3D samples instead of the real 3D microstructures. The input parameters for the tensile-property simulations were measured via atomic force microscopy, which is used to measure the phase stiffness on a sub-micro scale. The results from these simulations were then compared with those of the experiment on a macro scale to confirm the correlation between the mechanical properties measured on the sub-micron and macro scales.

Thermo-mechanical stress analyses of solid oxide fuel cell anode based on three-dimensional microstructure reconstruction

During high-temperature operation of a solid oxide fuel cell, the stresses caused by mismatch of thermal expansion coefficients between different materials and external mechanical loads may cause the rising of damage risk of nickel-yttria-stabilized zirconia anode. It is quite difficult to quantify the mechanical characteristics of a composite anode without investigating on the stress distribution in its real microstructure. However, the high operating temperature and extremely complex microstructure in micro-scale determine the high difficulty in in-situ measurement of thermo-mechanical stress distribution. In this work, the microstructures of six different anode samples, fabricated by using identical nickel oxide-yttria-stabilized zirconia powder mixture, are reconstructed in three-dimension based on the dual-beam focused-ion-beam-scanning-electron-microscopy. The three-dimensional thermo-mechanical stress distributions of different microstructures are conducted at operating temperature based on the finite element method. The effects of both thermal expansion coefficients mismatch between nickel and yttria-stabilized zirconia and external mechanical loads are analyzed. The mechanical failure probabilities of yttria-stabilized zirconia phase in different reconstructions are estimated based on the obtained stress distributions to investigate the influence of microstructure characterizations on nickel-yttria-stabilized zirconia anode strength.

Efficacy and efficiency of dual electrochemical strategies for remedying rebar corrosion in heterogeneous concrete

Corrosion of rebars continues to be a major concern for the durability of steel reinforced concrete structures. To minimize the risk involved with rebar corrosion in this study two electrochemical remedial strategies, i.e., electrical injection of corrosion inhibitors (EICI) and electrochemical removal of chlorides (ERC), are investigated in this study using an integrated analytical and experimental approach. A mathematical model in the form of a multiphysics, non-linear differential equation was established for evaluating the efficacy and efficiency of a parallel EICI and ERC scheme, aided by three accelerated electrochemical experiments developed for parametric determination and for model validation. Based on the method of three-dimensional microstructure reconstruction, a finite element analysis was applied to solve the mathematical model that couples three key mechanisms of mass transport in the matrix of heterogeneous concrete, including natural diffusion, hydraulic migration and electromigration. The validated mathematical model indicates an optimum current density of 3 A/m2 for implementation. Being able to predict the concentrations of multiple chemical species at arbitrary times and locations in heterogeneous concrete, the developed mathematical model demonstrates to be a powerful tool for evaluating general chemical-deterioration problems and remedial strategies for concrete structures.

Evaluation of La0.57Sr0.38Co0.2Fe0.8O3-δ Electrode Performance Degradation Based on Three-Dimensional Microstructure Reconstruction and Electrochemical Simulation

Time evolutions of electrochemical performance of La0.57Sr0.38Co0.2Fe0.8O3-δ (LSCF) during operation are evaluated based on three-dimensional microstructural analysis and electrochemical simulation. After 100 hours of operation at 700 and 800°C, increases in polarization and ohmic resistances are observed. Segregated Sr phase is found inside the LSCF layer after the operation at 800°C. LSCF, GDC, segregated Co phase, segregated Sr phase, GDC interlayer, interdiffusion phase, SZO phase, and YSZ electrolyte are segmented and reconstructed by focused ion beam – scanning electron microscopy (FIB-SEM) to analyze their microstructural characteristics in detail. No significant changes are observed for double phase boundary and triple phase boundary densities. Electrochemical simulations using the three-dimensional structures including all phases are carried out to predict the degradation factor. The increase in polarization resistance is well reproduced by assuming a decrease in the surface exchange rate, On the other hand, the increase in the ohmic resistance is attributed to the constriction of the ionic current in the inter-diffusion layer.

Modelling of thermal expansion of single- and two-phase ceramic polycrystals utilising synthetic 3D microstructures

Determining effective properties of heterogeneous materials is interesting not only in the theoretical aspect but also in the practical one – engineering or technological. Therefore, steps allowing to obtain values of the effective coefficients of thermal expansion of single- and two-phase ceramic polycrystals, were described in the hereby work. At first, based on scanning electron microscope (SEM) images of materials used in the research, the basic microstructure parameters were determined. Then, by means of the Scheil-Schwartz-Saltykov method, the distributions of equivalent grain diameters (3D) were established. The obtained information on grains morphology were utilised in the reconstruction of three-dimensional microstructures of analysed polycrystals. Reconstructions were performed by means of the Laguerre tessellation. In the next step, on the basis of the proposed criterion, numbers of independent realisations – necessary to obtain effective coefficients of thermal expansion – were determined. Finally, the planned simulations (Finite Element Analysis – FEA) were performed and the obtained effective values were compared with the dilatometric measurements results. The path, presented in this work, leading from two-dimensional microstructure images via synthetic three-dimensional models of polycrystals to FEA models, allowed to obtain the effective coefficients of thermal expansion within the measurement error of experimental values.

Three-Dimensional Microstructural Characterization of Cast Iron Alloys for Numerical Analyses

In this paper, we aim at characterizing three different cast iron alloys and their microstructural features, namely lamellar, compacted and nodular graphite iron. The characterization of microscopic features is essential for the development of methods to optimize the behavior of cast iron alloys; e.g. maximize thermal dissipation and/or maximize ductility while maintaining strength. The variation of these properties is commonly analyzed by metallography on two-dimensional representations of the alloy. However, more precise estimates of the morphologies and material characteristics is obtained by three-dimensional reconstruction of microstructures. The use of X-ray microtomography provides an excellent tool to generate high resolution three-dimensional microstructure images. The characteristics of the graphite constituent in the microstructure, including the size, shape and connectivity, were analyzed for the different cast iron alloys. It was observed that the lamellar and compacted graphite iron alloys have relatively large connected graphite morphologies, as opposed to ductile iron where the graphite is present as nodules. The results of the characterization for the different alloys were ultimately used to generate finite element models.

Three-Dimensional Microstructure Reconstruction and Phase Boundary by Serial Sectioning and Numerical Approach

The microstructure and pores of the consolidated powder materials are important to determine the final properties of the pieces produced. Therefore, the control of their volume fractions and morphologies represent actual technological challenges. Besides the quantification of the phase and pores volume fractions there are important features related with the morphological aspects and the phase boundary structure that reveals the mechanisms of properties improvements, which are crucial for development of materials obtained by powder metallurgy techniques. In this paper, stereological measurements of pores present in the material stainless steel (316L) are obtained by combining serial sectioning technique and optical microscopy (OM). Dilatometric experiments were carried out in order to simulate heat treatment for sintering the material. The 3D microstructures are obtained by images reconstruction using numerical pixels treatment and stereological results of the porosities are presented.

Development of novel three-dimensional reconstruction method for porous media for polymer electrolyte fuel cells using focused ion beam-scanning electron microscope tomography

Abstract An accurate method for three-dimensional reconstruction of microstructures is developed to analyse mass transport phenomena in microporous layers (MPLs). The proposed method involves the use of focused ion beam–scanning electron microscope tomography. Numerical calculations of the structural and mass transport properties of MPLs are conducted using the reconstruction results and these results are found to be in good agreement with experimental results, demonstrating the accuracy of the new method.

Three-dimensional Microstructure Reconstruction of Pleated Air Filter Media and Simulation of Its Performance in Dust Filtration Stage

Three-dimensional Microstructure Reconstruction of Pleated Air Filter Media and Simulation of Its Performance in Dust Filtration Stage

3D reconstruction size effect on the quantification of solid oxide fuel cell nickel–yttria-stabilized-zirconia anode microstructural information using scanning electron microscopy-focused ion beam technique

Self-made conventional nickel–yttria-stabilized zirconia composite anodes after reduction and 500h operation were analyzed by three-dimensional microstructure reconstruction based on focused ion beam-scanning electron microscopy technique. Interfacial area, three-phases-boundary density and phase volume fraction were measured based on the three-dimensional microstructure reconstruction to quantitatively study the statistical characterization of solid oxide fuel cell nickel–yttria-stabilized-zirconia anode microstructure before and after operation. It is found that for anode operated with long time, it is necessary to increase the corresponding three-dimensional reconstruction size to suppress the influence of microstructure variation caused by Ni agglomeration in order to obtain more accurate microstructural quantification information.

Micron-scale phenotyping quantification and three-dimensional microstructure reconstruction of vascular bundles within maize stalks based on micro-CT scanning.

Vascular bundles within maize (Zea mays L.) stalks play a key role in the mechanical support of plant architecture as well as in water and nutrient transportation. Convenient and accurate phenotyping of vascular bundles may help phenotypic identification of germplasm resources for breeding. Based on practical sample preparation procedures for maize stalks, we acquired serials of cross-sectional images using a micro-computed tomography (CT) imaging device. An image processing pipeline dedicated to the phenotyping of vascular bundles was also developed to automatically segment and validate vascular bundles from the cross-sectional images of maize stalks, from which phenotypic traits of vascular bundles, i.e. number, area, and spatial distribution, were calculated. More profound quantification of spatial distribution was given as area ratio of vascular bundles, which described the distribution of vascular bundles associated with the centroid of maize stalks. In addition, three-dimensional visualisation was performed to reveal the spatial configuration and distribution of vascular bundles. The proposed method significantly improves computation accuracy for the phenotypic traits of vascular bundles compared with previous methods, and is expected to be useful for illustrating relationships between phenotypic traits of vascular bundles and their function.

Three-Dimensional Microstructure Reconstruction and Finite Element Simulation of Gas Pores in the High-Pressure Die-Casting AZ91 Mg Alloy.

High-pressure die-casting (HPDC) AZ91 tensile specimens were used to investigate characteristics of gas pores and their effects on mechanical properties of HPDC AZ91 magnesium (Mg) alloy. Combining the stereoscopic morphology of gas pores obtained from a three-dimensional (3D) reconstruction technique with the experimental data from uniaxial tensile testing, we worked on finite element simulation to find the relationship between gas pores and the mechanical properties of HPDC AZ91 Mg alloy. Results indicate that the 2D metallography images have one-sidedness. Moreover, gas pores >100 µm in the center region have a remarkable negative influence on the ultimate tensile strength (UTS) and elongation. With an increase in the size of large gas pores in the center region, the UTS and elongation of the material decreases. In addition, the distribution of gas pores in the specimens and the areal fraction of gas pores >100 µm on cross sections can also affect the UTS and elongation to some extent.

Three-dimensional reconstruction of compacted graphite in vermicular cast iron by manual serial sectioning

Microstructural characterization is an important tool to optimize the properties of engineering materials. Metallography is a common technique, but it provides two-dimensional images from (sometimes complex) three-dimensional microstructures. Three-dimensional reconstruction of microstructures provides more precise informa tion on the morphologies of the material microconstituents, providing more accurate control over their physical and mechanical properties. The main barrier for the dis semination of this technique is the extensive time dispended to obtain a record of in numerous metallographic polished planes and the 3D image reconstruction procedure. The present work shows a practical example of three-dimensional reconstruction of the graphite phase in a compacted graphite cast iron executed by manual technique and using optical microscopy.

Grain boundary character distribution derived from three-dimensional microstructure reconstruction

Manual serial sectioning which includes consecutive steps of sample preparation and Electron Back Scattering Diffraction (EBSD) measurement was employed to extract the twodimensional (2D) sections of a pure nickel sample and to reconstruct the three-dimensional (3D) microstructure. A general alignment algorithm based on the minimization of misorientation between two adjacent sections has been developed to accurately align the sections. By employing this alignment algorithm, any in-plane (translational) and rotational misalignment as well as the planparallelity can be corrected. Surface triangulation technique was used to reconstruct the grain boundary surfaces. The Grain Boundary Character Distribution (GBCD) was derived from reconstructed grain boundaries. The results show that a smoother grain boundary plane can be obtained after precise translational and rotational alignment and correction of planparallelity.The relative grain boundary energy was computed as a function of the five grain boundary parameters based on equilibrium at triple lines. The results show that the grain boundary planes carrying a Σ3 type misorientation are dominantly parallel to the {111} crystal plane, which indicates the presence of coherent twin boundaries. It was observed that coherent Σ3 type boundaries exhibit the minimum relative grain boundary energy, which is approximately 57% smaller than the average of all Σ3 boundaries, including also incoherent twin boundaries.

Investigation of slice thickness and shape milled by a focused ion beam for three-dimensional reconstruction of microstructures

Three-dimensional reconstructions of microstructures produced by focused ion beam (FIB) milling usually assume a uniform slice thickness with flat and parallel surfaces. Measurement of the actual slice thickness and profile is difficult, and is often simply ignored. This paper reports the use of artificial 3D structures of known geometry to enable the full 3D profile of a sequence of slices produced by FIB to be measured for the first time. A transient period at the beginning of a milling process is observed in which the actual slice thickness varies by as much as ±50% from the target thickness (with significantly greater error near the base of the slice), before settling to a ±20% variation as the milling progresses. Although SEM images appear to show flat milled surfaces perpendicular to the top surface, the development of a curved, tapering milled surface is also observed. This profile is then maintained through the milling process with the bottom of the slice lagging the top by up to three slice thicknesses.