Please use this identifier to cite or link to this item:
|Title:||Investigation of microstructure and mechanical properties of A131 EH36 steel additively manufactured by selective laser melting and direct laser deposition||Authors:||Wang, Jingjing||Keywords:||Engineering::Materials::Metallic materials::Alloys||Issue Date:||2020||Publisher:||Nanyang Technological University||Source:||Wang, J. (2020). Investigation of microstructure and mechanical properties of A131 EH36 steel additively manufactured by selective laser melting and direct laser deposition. Doctoral thesis, Nanyang Technological University, Singapore.||Abstract:||The microstructure of selective laser melting (SLM) processed ASTM A131 EH36 high strength low alloy steel has been studied in-depth . However, there has been no report on the fatigue properties of SLM processed EH36 steel in the literature. The research on direct laser deposited (DLD) EH36 steel has not been reported either. In this research, ASTM A131 EH36 high strength low alloy steel was developed through two additive manufacturing (AM) processes, i.e. selective laser melting (SLM) and direct laser deposition (DLD), using EH36 steel powder. The research included the optimization of 3D printing process and characterization of the microstructures and mechanical properties of the additively manufactured (AMed) EH36 steel samples. The SLM process parameters were optimized through tuning laser scanning speed from 100 to 400 mm/s, hatch spacing from 0.08 to 0.13 mm, laser power from 145 to 240 W, and preheating temperature from 100 to 200°C with a layer thickness of about 50 μm under a chessboard scanning strategy. It was found that increasing laser power and preheat temperature and lowering scanning speed improved the densification of the built samples to near full density (>99%). With lower laser scanning speeds, the grains were coarser and more like a cellular structure while with higher scanning speeds the grains were finer and more like a cellular-dendritic structure in the samples (around 1 μm). A large plate-like martensitic structure was found in the coarse columnar grains (up to 20 μm). However, different morphologies of the martensite grains were found mainly in cellular, columnar and lath shapes, which probably resulted from different transformation mechanisms and were controlled by cooling rate. Electron back scattered diffraction analysis showed that the prior elongated columnar grains with the martensitic plate-like structure delineated the columnar grain boundaries. The fractured surfaces of the printed samples, examined with scanning electron microscopy, showed mostly a dimple type of failure. The microsegregation behavior of the samples and the degree of element escaping from the lattices were attributed to this rapid solidification process. The high cycle fatigue behavior, especially fatigue life, of the samples built at lower scanning speeds excelled those built at higher scanning speeds at which porosity induced fatigue failure dominated. S-N curves were built from the experimental data with respect to different scanning speeds and the fatigue failure mechanism was proposed. The EH36 steel samples were printed using DLD process in four different orientations, namely, horizontal 0° (XY_0°) & 45° (XY_45°) and vertical 45° (XZ_45°) & 90° (XZ_90°). The results showed that all the steel samples printed in the four orientations satisfied the ASTM standards for tensile and charpy impact properties. The microstructure and mechanical properties of the samples could be tuned by controlling the printing process parameters. The different thermal cycles encountered in the different build orientations could also be complemented by adjusting the process parameters. The lack of fusion defects were more prominent in the samples built in the two vertical directions, which was probably due to oxidation that impacted the interlayer bonding. The fatigue life was affected most severely by the lack of fusion defects for the samples printed in the direction XZ_90°. The fatigue fracture mechanism of the DLD printed EH36 steel samples was proposed and the mechanical properties were closely correlated with the microstructure. In addition, the influence of surface porosity on the fatigue life of AMed EH36 steel samples via selective laser melting (SLM) was studied in-depth by using a numerical method, which the pores were simulative AM 3D pores. The surface pores in both simplified semi-ellipsoid/spheroid and more realistic triangular shapes were evaluated in the first place. Stress concentration factor Kt analyzed by ABAQUS and fatigue life assessed by FE-SAFE were closely related to the AM 3D pore size, shape, position, orientation as well as their interactions, which confirmed their effects on the crack initiation stage during fatigue testing. The effectiveness of adding a surface finish factor Kt to a smooth surface fatigue model was subjected to internal pore geometry in terms of Kt values ranging from 1 to 5.5. Different algorithms were attempted with the stress-life method gives a higher estimation than a strain-life method while became invalid when the stress concentration situation was rather serious (around Kt = 2.5). There was no much difference between the uniaxial and multi-axial fatigue algorithms because the uniaxial tensile cyclic loading was applied to the model. Theoretically calculated total fatigue life (crack propagation) using linear elastic fracture mechanics (LEFM) fell in the same order (within a factor of 2) as the experimental results for all the samples. AMed material parameters were used for fatigue life prediction where the microstructure element was counted. The pores in an AMed material could severely deteriorate its mechanical properties, such as strength, Young’s modulus and fatigue lifetime, etc. The statistical analysis showed the scatter band of the SN curves with a factor of 8 with respect to different scanning speeds. All the above results contributed to scientific knowledge and provided a design guideline for possible marine and offshore applications of the AMed grade EH36 steel.||URI:||https://hdl.handle.net/10356/145851||Rights:||This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0).||Fulltext Permission:||open||Fulltext Availability:||With Fulltext|
|Appears in Collections:||MAE Theses|
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.