5 Werkstofftechnik
Filtern
Erscheinungsjahr
- 2021 (56) (entfernen)
Dokumenttyp
- Vortrag (56) (entfernen)
Sprache
- Englisch (56) (entfernen)
Referierte Publikation
- nein (56)
Schlagworte
- Corrosion (6)
- Microstructure (4)
- Ontology (4)
- Additive Manufacturing (3)
- Additive manufacturing (3)
- Creep behavior (3)
- Thermoelectrics (3)
- Carbon capture storage (2)
- Ceramics (2)
- Chemically complex alloy (2)
Organisationseinheit der BAM
- 5 Werkstofftechnik (56) (entfernen)
Eingeladener Vortrag
- nein (56)
Laser powder bed fusion (LPBF) is an additive manufacturing process for materials which inherently tends to yield various degrees of metastable hierarchical microstructures, defects and high residual stresses in the as-built condition depending on the process parameters. The understanding of the evolution of these typical features during heat treatment and subsequent thermal and mechanical ageing is crucial for the wider acceptance for safety critical structures. A multi-disciplinary research project at BAM studying the development of the microstructure, defects, residual stresses typical of LPBF 316L and their evolution during thermal and mechanical ageing has led to insights into the stability of these inherent features. This presentation aims to give a broad overview of the project with a few specific cases of investigation. Firstly, the formation of residual stresses, the nature of the initial microstructure, the tensile properties and a modelling approach to understand the anisotropy will be presented. This will be followed by examples of studies of their evolution during heat treatment, long term thermal exposure, and room temperature and high temperature mechanical testing compared to a baseline of conventional wrought variant of the same alloy.
Ceramic multilayer technology is an attractive approach for the cost-effective fabrication of thermoelectric generators. Therefore, efforts are being made to co-sinter two promising thermoelectric oxides, namely calcium cobaltite and calcium manganate. In this study, calcium cobaltite, calcium manganate and release tapes were pressure assisted sintered. A major challenge here is the cracking of calcium manganate during cooling. A relationship between the properties of the release tape used in pressure-assisted sintering and the cracking behavior was observed experimentally.
To understand the origin of failure, formed reaction layers in the multilayer were analyzed by EDX and grazing incident XRD. Based on this analysis, bulk samples were prepared, and thermal expansion and Young's modulus and were determined thereon, if they were not known from the literature. The biaxial strength of the thermoelectric oxides was determined by the ball on three ball method.
The thermal stresses during cooling of different multilayer designs were estimated by finite element simulations. The stresses caused by the reaction layers turned out to be negligible. The FEM study indicated further, and a validation experiment proved, that the thickness of the release tape has the main effect on thermal stresses during cooling in single material. For best performance, the design of a thermoelectric multilayer generator needs to consider thermoelectric performance and thermal stresses during cooling.
CALM is software for determining the Bravais lattice type and the resulting lattice parameters from a single Kikuchi pattern. It requires the definition of 4 bands and a single bandwidth from which all other band positions as well as bandwidths are derived. For band detection, it uses the Funk transform, which allows detection of twice as many bands as usual. CALM works for any symmetry and requires low-noise patterns of at least 320x240 pixels. The resulting errors are <2% even for such small patterns, assuming good quality. The relative errors are <0.5%. However, this requires a projection centre position best derived from a sample of a cubic phase in CALM. However, this must have been recorded under identical conditions. Hundreds of Kikuchi patterns of phases with different symmetries were examined.
The Gibbs free energy of a grain boundary is a complex thermodynamic function of temperature, pressure, and composition. These complexities add to the intrinsic crystallographic and chemical constraints imposed by the adjacent bulk phase. Recently we have proposed a density-based model for assessing grain boundary thermodynamics that enables a CALPHAD-informed description of the grain boundary. As such, the Gibbs free energy of the grain boundary is directly linked with available CALPHAD thermodynamic data. In this talk, new aspects of interfacial segregation and phase transformation are revealed by benchmarking the current model for various experimental cases, including several steels, high-entropy alloys and aluminum alloys. The effects of elastic interactions on the grain boundary segregation and the application of the model to a nanocrystalline Pt-Au alloy, with numerous grain boundaries of various characters, will be discussed.
The following work deals with the quantitative fracture surface evaluation in damage analysis. So far, fracture surfaces have almost exclusively been evaluated qualitatively, i.e. the presence of fracture features is documented and their surface proportions are estimated, if necessary. Many years of experience are required, as well as an intensive comparison with defined comparative images from the literature. The aim of this work is the development of classifiers which can recognize fracture mechanisms or fracture features in scanning electron microscope images (SEM). The basis is 46 SEM images, which have been evaluated by fractography experts with regard to fracture features. The existing data set of images is expanded using augmentation methods in order to increase the variability of the data and counteract overfitting. Only convolutional neural networks (CNN) are used to create the classifiers. Various network configurations are tested, with the SegNet achieving the best results.
Reliable characterization of materials at the nanoscale regarding their physio-chemical properties is a challenging task, which is important when utilizing and designing nanoscale materials. Nanoscale materials pose a potential toxicological hazard to the environment and the human body. For this reason, the European Commission amended the REACH Regulation in 2018 to govern the classification of nanomaterials, relying on number-based distribution of the particle size.
Suitable methods exist for the granulometric characterization of monodisperse and ideally shaped nanoparticles. However, the evaluation of commercially available nanoscale powders is problematic. These powders tend to agglomerate, show a wide particle size distribution and are of irregular particle shape.
Zinc oxide, aluminum oxide and cerium oxide with particle sizes less than 100 nm were selected for the studies and different preparation methods were used comparatively.
First, the nanoparticles were dispersed in different dispersants and prepared on TEM-supported copper grids. Furthermore, individual powders were deposited on carbon-based self-adhesive pads. In addition, the samples were embedded by hot mounting and then ground and polished.
The prepared samples were investigated by scanning electron microscopy (including the transmission mode STEM-in-SEM) and Dynamic Light scattering. The software package ImageJ was used to segment the SEM images and obtain the particle sizes and shapes and finally the number-based particles size distribution with size expressed as various descriptors.
CO2 quality specifications are not only a matter of CO2 purity (i.e. CO2 content).
The “rest” also matters, in particular contents of reactive impurities affecting material corrosion (and rock alteration).
Also chemical reactions in CO2 stream needs to be considered, in particular when combining CO2 streams of different compositions.
To apply high-entropy alloys (HEA) of the CrMnFeCoNi family in challenging atmospheres, their degradation behavior under harsh environments needs to be investigated. Oxidation studies to HEAs have not been extensively investigated and most of them are concentrated on environments like synthetic air, laboratory air, CO/CO2, O2 and H2O atmospheres. Main corrosion products which were identified after aging times of up to 100 h are Mn2O3 (≤800°C) and Mn3O4 (≥800°C).
Another corrosive medium in high temperature applications is SO2, which preferentially forms sulfides on commercial steels for example. These can be occurred both in the oxide layer and at the oxide/metal interface. For instance, on Fe-Cr based alloys sulfides (Cr5S6) were detected along grain boundaries and their number increases with exposure time and Cr-content in the alloy. These sulfides show an increased hardness, compared to the bulk alloy, and cause an embrittlement of the grain boundaries. This is a serious material degradation phenomenon, now addressed for the case of HEAs.
In the present study metal sulfides were identified after corrosion of the HEA CrMnFeCoNi alloy in an Ar-0.5vol.%SO2 atmosphere at 800°C for 24 h, 48 h, 96 h and 192 h exposure time. After all three duration times, a thin non-protective Cr2O3 layer has formed at the oxide/alloy interface. At the gas side a thick Mn3O4 layer with local voids containing sulfur could be detected by SEM-EDS analysis. Furthermore, S precipitates could be detected in the bulk material near the surface. These sulfides were characterized in detail by scanning and transmission electron microscopy. Based on these results, a model for grain boundary sulfidation of high-entropy alloy CrMnFeCoNi is discussed.
This presentation shows some experimental results of the characterization of the creep behavior of LPBF 316L, which has been poorly studied and understood to date. The presentation includes results regarding the mechanical properties, the initial microstructural state and its evolution under loading, and the damage mechanism. This work was done within the BAM focus area materials project AGIL. As a benchmark to assess the material properties of the LPBF 316L, a conventionally manufactured variant was also tested.
Porosity in single-crystal nickel-base superalloys is removed by hot isostatic pressing (HIP) at temperatures above gamma’-solvus where the material is very soft and ductile. For example, single-crystal nickel-base superalloy CMSX-4 is HIPed at temperature 1288 °C, which is slightly higher than the gamma’-solvus temperature of this alloy equal to about 1280 °C. It is assumed that pore shrinking during HIP is mostly due to dislocation creep. Such a modelling of HIP of CMSX-4 was started in our group on the base of results of creep tests of [001] single-crystals at 1288 °C [1]. However, it was found later [2] that the alloy CMSX-4 shows very strong creep anisotropy at 1288 °C. Therefore, for calibration of the creep law, creep tests of different orientations under different stress levels are required at the HIP temperature. This was the main task of present work. Single-crystals of CMSX-4 of axial orientations [001], [011], [123] and [111] were cast by VIAM Moscow and tested by BAM Berlin under creep conditions at 1288 °C and stress levels between 4 MPa and 16 MPa. At all stress levels, the creep rate increases by an order of magnitude when changing the orientation from [001] to [111] with [011] and [123] orientations in between. Such a character of creep anisotropy corresponds to the orientation dependence of the Schmid factor for octahedral glide.
The crystal viscoplasticity model developed in [1] was improved to better represent the time induced softening observed during creep. The creep tests for different stresses and orientations as well as pore closure were simulated. The results of pore closure simulation are compared with measurements of porosity decrease during Hiping.