2D Scanning Electron Microscopy Images Research Articles

Experimental and Finite Element Simulation of Polyolefin Elastomer Foams Using Real 3D Structures: Effect of Foaming Agent Content.

In this study, polyolefin elastomer (POE) foams were prepared without any curing agent using a single-step foaming technique. The effect of azodicarbonamide (ADC) content as a chemical foaming agent on the foams' morphology and mechanical properties was studied using scanning electron microscopy (SEM), mechanical properties (tension and compression) and hardness. The results showed that increasing the ADC content from 2 to 3, 4 and 5 phr (parts per hundred rubber) decreased the foam density from 0.75 to 0.71, 0.65 and 0.61 g/cm3, respectively. The morphological analysis revealed that increasing the ADC content from 2 to 4 phr produced smaller cell sizes from 153 to 109 µm (29% lower), but a higher cell density from 103 to 591 cells/mm3 (470% higher). However, using 5 phr of ADC led to a larger cell size (148 µm) and lower cell density (483 cells/mm3) due to cell coalescence. The tensile modulus, strength at break, elongation and hardness properties continuously decreased by 28%, 21%, 16% and 14%, respectively, with increasing ADC content (2 to 5 phr). On the other hand, the compressive properties, including elastic modulus and compressive strength, increased by 20% and 64%, respectively, with increasing ADC content (2 to 5 phr). The tensile and compression tests revealed that the former is more dependent on foam density (foaming ratio), while the latter is mainly controlled by the cellular structure (cell size, cell density and internal gas pressure). In addition, 2D SEM images were used to simulate the foams' real 3D structure, which was used in finite element methods (FEM) to simulate the stress-strain behavior of the samples at two levels: micro-scale and macro-scale. Finally, the FEM results were compared to the experimental data. Based on the information obtained, a good agreement between the macro-scale stress-strain behavior generated by the FEM simulations and experimental data was obtained. While the FEM results showed that the sample with 3 phr of ADC had the lowest micro-scale stress, the sample with 5 phr had the highest micro-scale stress due to smaller and larger cell sizes, respectively.

Digital transformation in rock physics: Deep learning and data fusion

Rock physics plays an essential role in geophysical reservoir characterization. It aims to build a bridge between geophysical measurements and in-situ rock and fluid properties. With the advancement of microscopic imaging and computer science, rock physics is transitioning to the digital age. This is referred to as digital rock physics (DRP). DRP provides a nondestructive and efficient way to determine physical rock properties directly from digital images. Over the last decades, it has become a routine tool in reservoir characterization by complementing or replacing expensive and time-consuming laboratory measurements. With the emergence of deep learning, DRP has advanced significantly from image processing to physical simulation. This paper presents an application of deep learning in multiscale fusion of digital rock images. It aims to overcome the trade-off between image resolution and field of view (FoV) by integrating imaging data from multiple sources including (1) 3D microcomputed tomography images at micronscale with a large FoV and (2) 2D scanning electron microscopy images at nanoscale with a small FoV. The reconstructed image integrates information of microstructures at different scales and helps characterize heterogeneous porous rocks more accurately. It is helpful to improve the prediction accuracy of effective rock properties and to have deeper insight into physical processes at pore scale. Data fusion based on deep learning would unlock new pathways for geophysical characterization of porous rocks, with broad implications for various subsurface applications such as groundwater transport, enhanced oil recovery, and geologic carbon sequestration.

Statistical Reconstruction of 3D Paper Structure Using Simulated Annealing Algorithm Based on 2D Scanning Electron Microscopy Image

ABSTRACT The microstructure of fibrous paper plays an important role in its property investigation. In this study, an approach is proposed to extrapolate a 2D image into a virtual 3D microstructure. Five types of handsheets made of different pulps were prepared. Then, a hybrid function of two-point correlation and lineal-path function (S 2 &L 2) and co-occurrence correlation functions (CCFs) was used in the simulated annealing reconstruction method. Thus, microstructures of two-phase fiber-pore handsheets were reconstructed using 2D scanning electron microscopy images. Finally, penetration simulations and calculations of the absolute permeability of handsheets were conducted. The statistical values of two-point correlation function (S 2) and lineal-path function (L 2) extracted from the reconstructed images were used to characterize the reconstruction accuracy, and the comparisons of reconstruction accuracy and time were made. The study results showed that the 3D microstructures of fibrous handsheets could be reconstructed effectively by S 2 &L 2 and CCFs, identifying with the targets. The accuracies were around 10− 5, and the reconstruction times by CCFs were shortened by 30–60% compared with S 2 &L 2. Moreover, the visual permeability simulation results could reflect the structural difference of handsheets, according to the calculated absolute permeability. These findings provide a guidance for 3D reconstruction of natural fiber paper.

Characterizing clay textures and their impact on the reservoir using deep learning and Lattice-Boltzmann simulation applied to SEM images

The presence of clays in hydrocarbon reservoirs challenges the producible amount of oil and gas significantly. Therefore, this study reports a detailed quantitative characterization of clays' specific properties from two fundamental aspects which include clays' type and amount, and their impact on reservoir's fluid flow. We used Scanning Electron Microscopy (SEM) images and respectively adopted deep learning for typing and quantifying clays, and the Lattice-Boltzmann Method (LBM) for flow simulations with and without the presence of clays. The trained deep learning model of the present study was translated into a MATLAB application that is a convenient tool for clay characterization by the future user. This model was trained using 2160 images of different clay minerals based on transfer learning using AlexNet and resulted in more than 95.4% accuracy while applied on the unforeseen images. Moreover, we established the technique of depth-slicing of 2D SEM images, which provides the possibility of 3D processing of the routine SEM images. The results from this technique proved that clays could reduce reservoir porosity and permeability by more than 30% and 400 mD, respectively. The introduced approach of the present study provides new insights into the detailed impacts of clay minerals on the reservoir's quality.

Descriptor-based method combined with partition to reconstruct three-dimensional complex microstructures.

A descriptor-based method combined with a partition approach is proposed to reconstruct three-dimensional (3D) microstructures based on a set of two-dimensional (2D) scanning electron microscopy (SEM) images. The features in the SEM images are identified and partitioned into small features using the watershed algorithm. The watershed algorithm first finds the local gray-level maxima, and partitions the features through the gray-level local minima. The 3D size distribution and radial distribution of the small spherical elements are inferred, respectively, based on the 2D size distribution and radial distribution using stereological analysis. The 3D microstructures are reconstructed by matching the inferred size distribution and radial distribution through a simulated annealing-based procedure. Combining with the proposed partition approach, the descriptor-based method can be applied to complex microstructures and the computational efficiency of the reconstruction can be largely improved. A case study is presented using a set of 2D SEM images with nanoscale pore structure from the low-density CSH (calcium silicate hydrate) phase of a hardened cement paste. Cross sections were randomly selected from the reconstructed 3D microstructure and compared with the original SEM images using the pore descriptors and the two-point correlation function with satisfactory agreement. Using the 3D reconstructed model, the properties of the sample material can be investigated on such a small scale as demonstrated in this paper on quantifying the absolute permeability.

RockFlow: Fast Generation of Synthetic Source Rock Images Using Generative Flow Models

Image-based evaluation methods are a valuable tool for source rock characterization. The time and resources needed to obtain images has spurred development of machine-learning generative models to create synthetic images of pore structure and rock fabric from limited image data. While generative models have shown success, existing methods for generating 3D volumes from 2D training images are restricted to binary images and grayscale volume generation requires 3D training data. Shale characterization relies on 2D imaging techniques such as scanning electron microscopy (SEM), and grayscale values carry important information about porosity, kerogen content, and mineral composition of the shale. Here, we introduce RockFlow, a method based on generative flow models that creates grayscale volumes from 2D training data. We apply RockFlow to baseline binary micro-CT image volumes and compare performance to a previously proposed model. We also show the extension of our model to 2D grayscale data by generating grayscale image volumes from 2D SEM and dual modality nanoscale shale images. The results show that our method underestimates the porosity and surface area on the binary baseline datasets but is able to generate realistic grayscale image volumes for shales. With improved binary data preprocessing, we believe that our model is capable of generating synthetic porous media volumes for a very broad class of rocks from shale to carbonates to sandstone.

3D pore system reconstruction using nano-scale 2D SEM images and pore size distribution analysis for intermediate rank coal matrix

This paper comprises the analysis of scanning electron microscopy (SEM) images, nuclear magnetic resonance (NMR) and mercury injection capillary pressure (MICP) data to quantify the pore distribution in coal matrix. We first generate the 3D pore system in the coal matrix based on the statistics of pore distribution obtained from 2D SEM images, and then extract the pore network using the maximal ball method. The influence of the reconstructed cube size and the 2D image resolution on the accuracy of the reconstructed 3D coal sample was analysed when generate the 3D digital coal smaple. It was observed that the highest resolution which was achieved for the studied samples (6 nm) resulted in the underestimation of porosity of the studied sample, and it is recommended for future to create several models with different resolution to find the most representative model, instead of apriori using the highest possible resolution. The extracted pore network was then used to analyse pore size distribution and perform capillary pressure simulation using pore network modeling. A comparison of the pore network analysis with NMR and measured MICP data demonstrated that the pore network extraction method simplified the results of distribution and underestimated the size of elongated pores and microfractures. The simulated and laboratory measured MICP shows significant difference partically bucease the network extraction method was not suitable for the studied samples and this could be overcomed by our future study of model MICP using direct simulation method in the reconstructed 3D model.

Microstructural Characterization of Ni/YSZ Electrodes in a Solid Oxide Electrolysis Stack Tested for 9000 Hours

The performance of solid oxide electrolysis cells (SOEC) is affected by microstructural changes inside the cell. In this work detailed post-mortem characterization is conducted on an SOEC stack tested for electrolysis of steam over 9000 h [1]. The stack, produced by Topsoe Fuel Cell A/S, consists of 25 identical repeating units connected in serial. Each repeating unit includes an SOEC cell and neighboring interconnects (ICs). The SOEC cell has a Ni-YSZ support layer and fuel electrode, a YSZ electrolyte, a LSCF-CGO (LSCF: (La,Sr)(Co,Fe)O3-δ, CGO:(Ce,Gd)O2) oxygen electrode and a CGO barrier layer between the electrolyte and the oxygen electrode. The objective of the present investigation is to identify modifications in the active Ni-YSZ electrode microstructure, which are believed to strongly affect the SOEC cell and stack performance. These microstructural changes are quantified thorough 2D and 3D microstructure analyses. Based on micrographs recorded for the 9 kh tested stack and a reference cell (i.e. a cell from the same production batch as those in the stack, but not long-term tested), a relative quantification of Ni, YSZ and pore phase fractions and particle sizes has been obtained through scanning electron microscopy (SEM) imaging analysis. The mean intercept length principle has been used to investigate 2D SEM images recorded on different cell locations. The ease of this type of characterization makes it possible to investigate several locations in the long term tested stack: both fuel electrode inlet and outlet, directly under the contact point between the IC and at the point where the cell is in contact with the gas channel. By comparing results of the reference and tested cell, significant changes in Ni and pore size are observed while the YSZ backbone does not show substantial changes. The 9000 h electrolysis testing results in a decrease in the Ni phase fraction and an increase in Ni particle size. Compared to the reference cell, the biggest microstructural change is observed at the location where the cell is directly in contact with the IC, where the Ni phase fraction is decreased from 28 % (reference) to 18 % and the Ni mean intercept length is increased from ~1 μm (reference) to ~1.4 μm. Based on the 2D image analysis locations of special interest were selected for 3D reconstruction of the electrode structure by focused-ion-beam (FIB) serial sectioning. The 3D data provides information on the true particle size distribution, avoiding the assumptions that need to be made for the 2D data. Besides, the 3D analysis allows the transport pathways to be characterized through each of the phase networks [2]. A comparison between 2D and 3D data is carried out to concisely investigated microstructural changes and relate them with the cell performance during long-term operation. [1] A. Brisse, J. Schefold, G. Corre, Q. Fu. Performance and Lifetime of Solid Oxide Electrolyzer Cells and Stacks. 11thEuropean SOFC & SOEC Forum, July 2014 [2] Peter Stanley Jørgensen, Søren Lyng Ebbehøj, and Anne Hauch. "Triple phase boundary specific pathway analysis for quantitative characterization of solid oxide cell electrode microstructure." Journal of Power Sources 279 (2015): 686-693. Figure 1

Quantitative characterization of cleavage and hydrogen-assisted quasi-cleavage fracture surfaces with the use of confocal laser scanning microscopy

“True” cleavage (TC) and quasi-cleavage (QC) fracture surfaces of low-carbon steel specimens tested in liquid nitrogen and after hydrogen charging respectively were investigated by quantitative confocal laser scanning microscopy (CLSM) and conventional scanning electron microscopy (SEM) with electron-backscattered diffraction (EBSD). Topological and crystallographic features of the TC fracture surface are found in good agreement with the generally accepted cleavage mechanism: TC facets diameters correspond to those of grains; the crack path strictly follows the crystallographic orientation of grains and the most of the cleavage cracks are parallel to {100} planes. On the 2D SEM images, the QC facets appeared resembling the TC ones in terms of river line patterns, shapes and sizes. However, the substantial differences between the topography of these two kinds of fracture surfaces were revealed by 3D CLSM: the average misorientation angle between QC facets and the roughness of the QC fracture surface were much lower than those measured for TC. It is demonstrated that all these features are attributed to the specific fracture mechanism operating during hydrogen-assisted cracking.

3D analysis of SEM images of corrosion casting using adaptive stereo matching

The corrosion casting method represents one of the most widely used technique to study the 3D microvascularization of many tissues, both in their normal and pathological conditions. For a long time this technique was used only to perform a qualitative evaluation of the images obtained by scanning electron microscopy (SEM). A quantitative evaluation of vascular parameters (e.g., interbranching and intervascular distances, angle measurements, lengths and diameters) was lacking, mainly because of the difficulties found in the measurement performed on 2D SEM images. Then, some authors reported a quantitative method based on the analyses of stereo-pair images that allowed precise morphometric measurements. To visualize the specimens in 3D, it was necessary to use red-green glasses. In this article we describe a new approach by which we can automatically obtain a 3D reconstruction of vascular cast specimen's surface directly from stereo-images. Moreover, we developed a software that performed micrometric measurements on the 3D construct generated from the stereo-pictures. In conclusion, implementing together these two softwares and applying them to corrosion casting samples made it possible to render in 3D the surface of corrosion cast as well as make quantitative measurements on the corrosion casts.