The aim of this work is to develop an automated procedure based on machine learning capabilities for the identification of the pearlite islands within the two-phase pearlitic–ferritic steel. The input parameters for the custom implementation of a braided neural network are provided as a data set of scanning electron microscopy images of metallographic specimens. The procedures related to the processing of the data and the optimization parameters affecting the final architecture and effectiveness of the network learning stage are examined. The objective is to find the best solution to the problem of ferritic–pearlitic microstructure segmentation, allowing further processing during, e.g., 3D reconstruction of data from serial sectioning. The work examines the various quality of input data and different U-Net architectures to find the one that can identify pearlite islands with the highest precision. Two types of images acquired from secondary electron (SE) and electron backscattered diffraction (EBSD) detectors are used during the investigation. The work revealed that the developed approach offers improvements in metallographic investigations by removing the requirement for expert knowledge for the interpretation of image data prior to further characterization. It has also been proven that artificial neural networks based on the deep learning process using extensible U-Net network architectures and nonlinear learning tools can identify pearlite islands within a two-phase microstructure, while the overtraining level remains low. Convolutional neural networks do not require manual feature extraction and are able to automatically find appropriate search functions to recognize pearlite structure areas in the training process without human intervention. It was shown that the network recognizes areas of analyzed steel with satisfactory precision of 79% for EBSD and 87% for SE images.
A two-dimensional mesoscale model based on the concept of hybrid cellular automata was used to study phase transformations in a complex phase steel during continuous cooling. This model enables simulation of the decomposition of austenite into ferrite, bainite, and martensite, accompanied by calculations of volume and grain boundary diffusion of carbon. In effect, as a result, one can observe the morphology of simulated microstructures, corresponding carbon segregation as well as microhardness distribution. These results with the kinetics of austenite to ferrite phase transformation and predicted values of the complex phase steel hardness are the subject of model validation. A series of dilatometric experiments were carried out with constant cooling rates in order to construct a CCT diagram and validate the presented model. The convergence of simulated results with empirical outcomes was confirmed quantitatively using a dedicated goal function and data summaries in the table and graphs. However, some qualitative and quantitative discrepancies in terms of microstructure morphology are indicated which was possible thanks to applying a wide range of different validation parameters of the model. It is emphasized how crucial is the use of appropriate validation methodology.
PL
Dwuwymiarowy mezoskalowy model oparty na koncepcji hybrydowych automatów komórkowych zastosowano do badania przemian fazowych w stali wielofazowej podczas ciągłego chłodzenia. Model ten umożliwia symulację rozpadu austenitu w ferryt, bainit i martenzyt wraz z obliczeniami objętościowej i granicznej dyfuzji węgla. W efekcie można zaobserwować morfologię symulowanych mikrostruktur, odpowiadającą im segregację węgla, a także rozkład mikrotwardości. Wyniki te wraz z kinetyką ferrytycznej przemiany fazowej i przewidywanymi wartościami twardości stali wielofazowej są przedmiotem walidacji modelu. Przeprowadzono szereg eksperymentów dylatometrycznych przy stałych szybkościach chłodzenia w celu opracowania wykresu CTPc i walidacji przedstawionego modelu. Zbieżność wyników symulacji z danymi empirycznymi została potwierdzona ilościowo za pomocą dedykowanej funkcji celu oraz zestawienia danych w tabeli i na wykresach. Jednakże, wskazano na pewne rozbieżności jakościowe i ilościowe pod względem morfologii mikrostruktury, co było możliwe dzięki zastosowaniu szerokiego wachlarza różnych parametrów do walidacji modelu. Podkreślono, jak istotne jest zastosowanie odpowiedniej metodologii walidacji.
The use of carbon fiber composites (CFC) for different applications is widespread. Carbon-based materials show, however, significant oxidative degradation in air. Modern materials are subjected to aggressive, corrosive environment. This type of environment may strongly reduce their mechanical properties. For the protection of CFC, it was necessary to apply coatings to the composite surface. In the presented paper, a chromium/chromium nitride (Cr/Cr2N) multilayer structure has been selected as the inner part. The outer part of the coating was a hydrogenated amorphous carbon (a-C:H), gradually implanted by Cr nanocrystals. The application of transmission electron microscopy (TEM) indicated that the proposed deposition method allowed the formation of a Cr/Cr2N multilayer of Λ = 150 nm, topped with a-C:H+ Cr23C6 composite of a varied carbides density. The micro-hardness of the deposited coatings was up to 14 GPa (at a load of 2 and 5 mN). The microstructure of the deposited coatings was described in detail by means of TEM in the authors’ recently published paper [1]. This paper is a continuation thereof, aimed at describing microstructure changes after a localized corrosion process. In order to study localized corrosion in coatings, particularly in metallic (Cr) interlayers, the potential measurements and voltammetry experiments were performed in a Ringer solution. The open-circuit potential reaches stable values after a sufficient time period. The results indicated that the presence of a-C:H+Cr23C6, the outer part of the coating, speeds up the localized corrosion process in Cr interlayers in the inner part of a coating.
The present study attempts to apply HE to 99.99% pure copper. The microstructure of the samples was investigated by both light microscopy and scanning transmission electron microscopy (STEM). Additionally, the microhardness was measured, the tensile test was made, and statistical analysis of the grains and subgrains was performed. Based on Kikuchi diffraction patterns, misorientation was determined. The obtained results show that microstructure of copper deformed by hydrostatic extrusion (HE) is rather inhomogeneous. The regions strongly deformed with high dislocation density exist near cells and grains/subgrains free of dislocations. The measurements of the grain size have revealed that the sample with an initial in annealed-state grain size of about 250 μm had this grain size reduced to below 0.35μm when it was deformed by HE to the strain ε=2.91. The microhardness and UTS are stable within the whole investigated range of deformation.
Processes of severe plastic deformation (SPD) are defined as a group of metalworking techniques in which a very large plastic strain is imposed on a bulk material in order to make an ultra-fine grained metal. The present study attempts to apply Equal-Channel Angular Pressing (ECAP), Hydrostatic Extrusion (HE) and combination of ECAP and HE to 99.5% pure aluminium. ECAP process was realized at room temperature for 16 passes through route Bc using a die having an angle of 90°. Hydrostatic extrusion process was performed with cumulative strain of 2.68 to attain finally wire diameter of d = 3 mm. The microstructure of the samples was investigated by means of transmission and scanning electron microscopy. Additionally, the microhardness was measured and statistical analysis of the grains and subgrains was performed. Based on Kikuchi diffraction patterns misorientation was determined. The measured grain/subgrain size show, that regardless the mode of deformation process (ECAP, HE or combination of ECAP and HE processes), grain size is maintained at a similar level – equal to d = 0.55-0.59 μm. A combination of ECAP and HE has achieved better properties than either single process and show to be a promising procedure for manufacturing bulk UFG aluminium.
6
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
The structure of Ceria doped Scandia Stabilized Zirconia (1Ce10ScSZ) electrolyte film deposited by EB-PVD (Electron Beam-Physical Vapour Deposition) technique on NiO-ZrO2 substrate was characterized by electron microscopy. The highly porous substrate was densely covered by deposited film without any spallation. The produced electrolyte layer was of a columnar structure with bushes, bundles of a diameter up to 30 ?m and diverse height. Between the columns, delamination cracks of few microns length were visible. The annealing of zirconia film at 1000 °C resulted in its densification. The columnar grains and delaminating cracks changed their shape into a bit rounded. High magnification studies revealed nanopores 5-60 nm formed along the boundaries of the columnar grains during annealing. High-quality contacts between the electrolyte film and anode substrate ensured good conductivity of the electrolyte film and high efficiency of SOFC.
This article presents a microstructure characterization of an alloy coating based on Ti-48Al-2Cr-2Nb- type γ + α2 intermetallic phases deposited via an out-of-pack aluminizing process. The goal of the aluminizing process was to obtain a coating composed of aluminum-rich TiAl2 or TiAl3 phases with greater oxidation resistance compared to the base alloy. The results showed that the gas-phase aluminizing process produced a coating with specific microstructural properties. The thickness of the layer obtained, including the transition zone, was approximately 20µm. X-ray diffraction (XRD) phase composition studies demonstrated that the outer coating zone was primarily composed of a TiAl2 phase, and its thickness was approximately 10 µm. Microanalysis of the chemical composition showed that, in addition to the main components, i.e. titanium and aluminum, chromium and niobium were present in the outer coating. Electron backscatter diffraction (EBSD) studies further indicated the probable presence of a TiAl2 phase. The coating obtained was of good quality, and cracks or pores, which are typical of coatings obtained via powder methods, were not detected.
PL
W artykule przedstawiono wyniki badań mikrostruktury warstwy wierzchniej stopu na osnowie faz międzymetalicznych γ + α2 typu Ti-48Al-2Cr-2Nb po procesie aluminiowania metodą out-of-pack. Celem procesu aluminiowania było uzyskanie warstwy zewnętrznej zbudowanej z bogatych w aluminium faz typu TiAi2 lub TiAl3 o wyższej odporności na utlenianie w porównaniu do stopu podłoża. Zrealizowane badania wykazały, że zastosowanie metody aluminiowania gazowego pozwoliło na wytworzenie na powierzchni stopu pokrycia o zakładanych właściwościach mikrostrukturalnych. Grubość otrzymanej warstwy, łącznie ze strefą przejściową wynosiła ok. 20 µm. Mikroanaliza składu chemicznego wykazała również, że poza głónymi składnikami tj. tytanem i aluminium w obszarze tym obecne były również chrom i niob. Badania EBSD wykazały ponadto prawdopodobną obecność fazy Ti3Al5. Uzyskane pokrycie było dobrej jakości, nie stwierdzono pęknięć ani pustek, typowych dla warstw otrzymanych metodami proszkowymi.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.