Ingenieurwissenschaften und zugeordnete Tätigkeiten
Filtern
Erscheinungsjahr
Dokumenttyp
- Dissertation (38) (entfernen)
Sprache
- Englisch (38) (entfernen)
Referierte Publikation
- nein (38)
Schlagworte
Organisationseinheit der BAM
- 8 Zerstörungsfreie Prüfung (5)
- 8.5 Röntgenbildgebung (3)
- 9 Komponentensicherheit (3)
- 5 Werkstofftechnik (2)
- 6 Materialchemie (2)
- 6.1 Oberflächen- und Dünnschichtanalyse (2)
- 8.1 Sensorik, mess- und prüftechnische Verfahren (2)
- 1 Analytische Chemie; Referenzmaterialien (1)
- 1.1 Anorganische Spurenanalytik (1)
- 5.2 Metallische Hochtemperaturwerkstoffe (1)
Additive Manufacturing (AM) in terms of laser powder-bed fusion (L-PBF) offers new prospects regarding the design of parts and enables therefore the production of lattice structures. These lattice structures shall be implemented in various industrial applications (e.g. gas turbines) for reasons of material savings or cooling
channels. However, internal defects, residual stress, and structural deviations from the nominal geometry are unavoidable.
In this work, the structural integrity of lattice structures manufactured by means of L-PBF was non-destructively investigated on a multiscale approach.
A workflow for quantitative 3D powder analysis in terms of particle size, particle shape, particle porosity, inter-particle distance and packing density was established.
Synchrotron computed tomography (CT) was used to correlate the packing
density with the particle size and particle shape. It was also observed that at least about 50% of the powder porosity was released during production of the struts.
Struts are the component of lattice structures and were investigated by means of laboratory CT. The focus was on the influence of the build angle on part porosity and surface quality. The surface topography analysis was advanced by the quantitative characterisation of re-entrant surface features. This characterisation was compared with conventional surface parameters showing their complementary
information, but also the need for AM specific surface parameters.
The mechanical behaviour of the lattice structure was investigated with in-situ CT under compression and successive digital volume correlation (DVC). The Deformation was found to be knot-dominated, and therefore the lattice folds unit cell layer wise.
The residual stress was determined experimentally for the first time in such lattice structures. Neutron diffraction was used for the non-destructive 3D stress investigation. The principal stress directions and values were determined in dependence of the number of measured directions. While a significant uni-axial stress state was found in the strut, a more hydrostatic stress state was found in the knot. In both cases, strut and knot, seven directions were at least needed to
find reliable principal stress directions.
Additive manufacturing (AM) offers significantly greater freedom of design compared to conventional manufacturing processes since the final parts are built layer by layer. This enables metal AM, also known as metal 3D printing, to be utilized for improving efficiency and functionality, for the production of parts with very complex geometries, and rapid prototyping. However, despite many technological advancements made in recent years, several challenges hinder the mass adoption of metal AM. One of these challenges is mechanical anisotropy which describes the dependency of material properties on the material orientation. Therefore, in this work, stainless steel 316L parts produced by laser-based powder bed fusion are used to isolate and understand the root cause of anisotropy in AM parts. Furthermore, an efficient and accurate multiscale numerical framework is presented for predicting the deformation behavior of actual AM parts on the macroscale undergoing large plastic deformations. Finally, a novel constitutive model for the plastic spin is formulated to capture the influence of the microstructure evolution on the material behavior on the macroscale.
The ability to reliably predict the amount and type of salts within a wet masonry is one of the most important and challenging subjects in the field of non-destructive combat of salt attack and preserving buildings and monuments of historical or archaeological value. To investigate the potential value of complex resistivity (CR) measurements for the early detection of salt and moisture related stone altering, a controlled salt type and concentration experiment series has been performed and a new procedure of an image based effective impedance modelling developed. Complex resistivity magnitude and phase measurements in the frequency range of 1 mHz to 100 Hz were acquired on a wide variety of wholly and partially brine saturated building stone samples. Deteriorating agents NaCl, Na2SO4, CaCl2 and MgSO4 were used. Since water and salt are not measured directly, the method is incapable of deducing possible building damages lacking a priori information. In the case of saturated materials, and provided thorough calibration data are available, a reliable estimate of salt concentration is possible from the measured resistivity magnitude. Additional indication on the dominant cation in the solution can be obtained from the measured resistivity phase, which is significantly higher the lower its valency. Furthermore, the measurements reported in this study give an important guide to the limitations of CR in obtaining pore surface area and pore throat estimates. For wet porous materials, in which polarization occurs due to complex surface conduction, the dominant pore throat and amount of specific surface affect its polarizability. Sandstones, sand-limestones and aerated concretes are more qualified observation objects (pore throats between 20 and 100 μm), whereas for bricks (pore sizes often < 5 μm) the method seems far less favorable. For those materials, which exhibit a Cole-Cole (C-C) type of relaxation, the phase peak is observed to decrease significantly with pore throat size and to occur at higher frequencies. The predicted power-law correlation between the C-C relaxation time and characteristic length scale (pore throat size for consolidated materials) is supported by the presented data. The experimental salinity study reveals how responsive polarization (in terms of imaginary conductivity) is to changing the ionic concentration or composition of the pore fluid. The properties of the electrical double layer and particularly its chemical composition are most likely the crucial controlling factors. The imaginary conductivity is observed to increase for most materials up to fluid salinities of about 1 S/m - a fact, that may be attributed to a mechanism of ion saturation within the electrical double layer; further ion supply seems to counteract this leading to ion-ion interactions, which decrease ionic mobility. CR measurements on partially saturated samples demonstrate the method’s sensitivity to water content. For most materials the imaginary conductivity component decreased significantly faster than the real component. Independent of the pore fluid and even though its salinity naturally increased during the evaporative drying procedure, the normalized saturation exponent of the quadrature component was about twice as high as the real one especially for clay-rich sandstones. The results revealed diverse behavior such as decreases and increases in relaxation time with saturation. In some cases a suppression of a distinctive relaxation curve at low saturations was observed. The results indicate, that contrary to conclusions from recent related studies, the correlation between a C-C relaxation time and hydraulic properties may be limited. In order to infer information on the charge distributions within the EDL, zeta-potentials and surface charge densities were obtained from electroacoustic measurements on particle suspension containing the same amounts and types of salt like used in the brine saturation CR study. The results add weight to the assumption that there is some universal positive relationship between zeta-potential and imaginary conductivity. A notable dependence of imaginary conductivity on diffuse layer surface charge was only observed in case of one sandstone (Cottaer), this being the material with the most abundant clay content. An image based effective impedance modelling approach revealed the fact that, even though it is not able to factor scale effects in, it is helpful to study general microstructural implications on CR responses. It was observed, that an increasing salt concentration (that creates a shrinking electrical double layer) reduces the peak phase and moves it towards higher frequency. The same effect would have an increasing fluid conductivity for a otherwise fixed microstructure. For some combinations of material and salt, however, in the experimental work the peak phase was observed to shift towards lower frequency: a phenomenon that could not be explained with the modelling. Therefore, it is assumed that chemical properties, like ion mobilities or other surface chemistry properties (ionexchange processes) must be regarded to conclusively explain surface conductivity mechanisms. Other experimental observations like the alteration of CR spectra at desaturation could likewise be modelled. If these models truly mimic the effective electrical properties, the results give new implications on the effective medium behavior. Concluding, the author values the complex resistivity method as a possible effective non-destructive testing (NDT) tool for a wide range of building stones. Depending on pore size and saturation important additional information can be obtained. In all cases, a priori information and calibration data are essential, that is, CR should not be treated as a stand-alone method. Further measurements are needed to develop a more complete model of the electrical double layer and its alteration with changing salinity and ion types.
In Europe a multi-national research project was initiated entitled “Rational Production and
Improved Working Environment through Using Self-Compacting Concrete”, followed by
another project entitled “Testing SCC”, which helped spreading the benefits of SCC to a wide
range of appliers. This project was also the basis of a widely accepted European guideline
on Self compacted concrete published by the European industry association bibm,
CEMBUREAU, EFCA, EFNARC, ERMCO, which again builds the basis of the European
standards for the testing of SCC (EN 12350, Parts 8 to 12) as well as for the actual
modernisation of the European concrete standard EN 206-1.
A new algorithm for detection of longitudinal crack-like indications in radiographic images is developed in this work. Conventional local detection techniques give unsatisfactory results for this task due to the low signal to noise ratio (SNR ~ 1) of crack-like indications in radiographic images. The usage of global features of crack-like indications provides the necessary noise resistance, but this is connected with prohibitive computational complexities of detection and difficulties in a formal description of the indication shape. Conventionally, the excessive computational complexity of the solution is reduced by usage of heuristics. The heuristics to be used, are selected on a trial and error basis, are problem dependent and do not guarantee the optimal solution. Not following this way is a distinctive feature of the algorithm developed here. Instead, a global characteristic of crack-like indication (the estimation function) is used, whose maximum in the space of all possible positions, lengths and shapes can be found exactly, i.e. without any heuristics. The proposed estimation function is defined as a sum of a posteriori information gains about hypothesis of indication presence in each point along the whole hypothetical indication. The gain in the information about hypothesis of indication presence results from the analysis of the underlying image in the local area. Such an estimation function is theoretically justified and exhibits a desirable behaviour on changing signals. The developed algorithm is implemented in the C++ programming language and testet on synthetic as well as on real images. It delivers good results (high correct detection rate by given false alarm rate) which are comparable to the performance of trained human inspectors.
Even though we often do not knowingly recognize them, nanoparticles are present these days in most areas of our daily life, including food and its packaging, medicine, pharmaceuticals, cosmetics, pigments as well as electronic products, such as computer screens. The majority of these particles exhibits a core-shell morphology either intendedly or unintendedly. For the purpose of practicability, this core-shell nanoparticle (CSNP) morphology is often assumed to be ideal, namely a spherical core fully encapsulated by a shell of homogeneous thickness with a sharp interface between core and shell material. It is furthermore widely presumed that all nanoparticles in the sample possess the same shell thickness. As a matter of fact, most real CSNPs deviate in several ways from this ideal model with quite often severe impact on how efficiently they perform in a specific application.
The topic of this cumulative PhD thesis is the accurate characterization of the actual morphology of CSNPs by advanced X-ray analytical techniques, namely X-ray photoelectron spectroscopy (XPS) and scanning transmission X-ray microscopy (STXM). A special focus is on CSNPs which deviate from an ideal core-shell morphology. In the paper from 2019 nanoparticle shell thicknesses are extracted from the elastic-peak intensities in an XPS spectrum based on an ideal particle morphology. This happens for a series of CSNP samples comprising a poly(tetrafluoroethylene) (PTFE) core and either a poly(methyl methacrylate) (PMMA) or polystyrene (PS) shell. The same paper as well as the paper from 2020 demonstrate for the first time, that the analysis of the inelastic background in an XPS spectrum of CSNPs can identify and quantify the heterogeneity of the shell and the incomplete encapsulation of the core.
The result from an XPS experiment is always an average across a large nanoparticle ensemble. Deviations from an ideal morphology within a single particle of the sample cannot be assessed separately. As opposed to that, a spatial resolution of 35 nm enables STXM to visualize the interior of single CSNPs which exhibit a sufficient X-ray absorption contrast between core and shell material. In the paper from 2018 a STXM analysis is demonstrated based on the example of the PTFE-PS CSNP samples already mentioned in the previous paragraph.
In the publication from 2021 (Ca/Sr)F₂ core-shell like nanoparticle ensembles for the practical use in, among others, antireflective coatings are investigated. These nanoparticles do not possess a sharp interface between core and shell material, which is why a shell thickness determination as described in the second paragraph is inappropriate. Instead, in-depth profiles of the chemical composition are obtained by XPS experiments based on synchrotron radiation with variable X-ray photon energy to elucidate the internal morphology of the particles. Additionally, theoretical in-depth profiles of Ca and Sr XPS peak intensities are simulated, in order to facilitate the interpretation of the experiments. Thus, an enrichment of CaF₂ at the particle surface was determined, which could hardly have been assessed by any other analytical technique. Because this kind of non-destructive depth profiling by XPS is very demanding, more than usual effort is spent on gapless documentation of the experiments to ensure full reproducibility.
Due to the vast diversity of nanoparticles differing in material, composition and shape, a measurement procedure cannot unalteredly be transferred from one sample to another. Nevertheless, because the papers in this thesis present a greater depth of reporting on the experiments than comparable publications, they constitute an important guidance for other scientists on how to obtain meaningful information about CSNPs from surface analysis.
Environmental Stress Cracking (ESC) and Slow Crack Growth (SCG) of PE-HD induced by external fluids
(2020)
High-density polyethylene (PE-HD) is widely used as a packaging material. Typical applications are pipes and containers for storage and transport of dangerous goods. For these applications, the understanding of the craze-crack damage mechanisms slow crack growth (SCG) and environmental stress cracking (ESC) is of importance. Since these mechanisms are considered to be the major causes of failure, their understanding is essential for inspection and release of those materials.
A well-established test method for the assessment of these damage mechanisms is the full-notch creep test (FNCT). It is used in this study for a detailed investigation of crack propagation phenomena in PE-HD container materials under the influence of different fluids such as air, water and aqueous detergent solutions (Arkopal N 100) as well as biodiesel and diesel. Based on the results of the FNCT, a classification scheme of different fluids is proposed, which allows for an assignment of the respective damage mechanisms. Hereby, it is differentiated between (i) inert, (ii) purely surface-active and (iii) additionally sorptive, bulk-active fluids with respect to SCG. If the test fluid changes the intrinsic properties (at the surface or in the bulk), the damage mechanism is addressed to ESC behavior.
In FNCT investigations, stress, temperature and specimen geometry were varied systematically. In addition to the time to failure as common measure for the resistance of a PE-HD type against crack propagation, specimen elongation was considered in detail.
Several imaging techniques were applied for fracture surface analysis of specimens tested in FNCT to gain novel information on SCG and ESC behavior. From height profiles obtained by laser scanning microscopy (LSM) and information on surface structures from scanning electron microscopy (SEM), indicators for the differentiation of the crack propagation mechanisms could be derived. Based on the LSM data, an algorithm for the distinction between ductile shear deformation and brittle crack growth as dominating failure mechanism was developed. Imaging techniques were also used for determination of crack propagation rates, which were related to time-resolved FNCT elongation data. From the time-resolved determination of crack lengths of partly damaged FNCT specimens, an increasing length of craze zone with a progressively propagating crack was revealed for the first time. This relation of crack and craze zones was specified by fracture mechanical considerations.
Glass is an amorphous material. When compared to steel, both its density and weight is three times lower. Its high theoretical strength makes it stand out as a premier material for a variety of applications. One such application is acting as a pressure resistant vessel for gas storage. Because glass has a high theoretical strength this makes it potentially suitable to withstand much higher pressures than steel or composite vessels. As a result of its brittle character, glass breaks when reaching a critical stress level. Therefore, the stress distribution during pressure load needs to be homogeneous without local stress peaks. At those peaks an initial crack will occur and the material will break. This PhD thesis is primarily concerned with the determination of the strength of several structures made of single hollow glass fibers during inner pressure treatment. Therefore, different kinds of hollow glass structures with varying parameters of shape and dimension were examined concerning their strength by determining the burst pressure. The burst pressure method was compared to the tensile test method, which poses the common test method for examining the strength of a material. The conclusion reached was that both test methods lead to comparable results and therefore, the burst pressure method poses an adequate tool for examining the strength of a hollow material against inner pressure. Another tool used in this thesis is the Finite Elements Method (FEM) simulation of internal stress and expansion of glass structures during pressure treatment. FEM was used to validate the burst pressure test results. A few selected material parameters needed to be incorporated, most notably the Young’s Modulus. Therefore, the expansion of single glass fibers was measured with light microscope during pressure load. Within the parameters of expansion, wall thickness and applied pressure, the Young’s Modulus was calculated with the Barlow’s Formula. According to the results, different two-dimensional models from single fibers to complex structures with up to 1000 single fibers were constructed and simulated with the CFD software Comsol Multiphysics. The expansion as well as the principal stress during pressure load was calculated. Different dimensions as well as different geometries of the glasses were considered to find a structure with the highest possible free volume and at the same time as less stress peaks as possible. This calculation was made in order to determine the best structure for gas storage. For this purpose the calculations were done with different dimensions of round single fibers right up to hexagonal structures consisting of more than one thousand round single fibers, which resulted in constant expansion of the structure. Furthermore, the problem of occurring interspaces between round single fibers, regarding their burst pressure-decreasing influence, was approached. Closing these interspaces with glass or other materials to avoid unsolicited pressure load led to increased strength of the structure and low storage capacities due to the increased weight and less free inner volume. The behavior of hexagonal fibers was determined as single fiber as well as in bundled condition. The walls between two hexagonal single fibers with applied inner pressure showed homogeneously distributed stress. Merely the outer walls without counter pressure showed high deformation and high structural stress. Based on that knowledge, several structures were modeled varying in different aspects. The fibers with hexagonal shape showed optimal stress distribution and high storage capacities because of high free inner volume, provided that these fibers are surrounded by additional fibers with identical inner pressure. Reducing the wall thickness for even higher free inner volume led to similar distribution but higher stress and expansion. To overcome the problem with the high stress at the outer fibers, the influence of outer fibers with different shape and dimension was simulated as well as the influence of solid glass fibers at the outer layer of the structure. The results showed that a structure with hexagonal thin-walled fibers should be surrounded by round fibers with higher wall thickness. This way the high stress peaks at the outer fibers are lowered. The examined practical strength of glass is about 100 to 1000 times lower than the theoretical strength. This is caused by defects, which may occur at the glass surface by handling or inside the material by defective production. Since the modeled results are based on the theoretical strength, the optimal wall thickness with a good compromise of strength and free inner volume needs to be found in practical tests. If further handling of the structures is necessary, an outer layer of solid fibers works as a protection layer against damages at the outer hollow glass fibers and increases the strength. Additionally, the influence of collapsing fibers inside a structure on the remaining system has been modeled as well as the influence of defects like holes or cracks at the surface or manufacturing induced defects inside the material. Any kind of defect leads to areas of high stress, whereby failure occurrence will be encouraged. In order to approve the theoretical results, the simulated structures were compared to the previously manufactured and tested ones. Due to the burst pressure test results, the tested structures showed low strength compared to the theoretical strength. This was primarily caused by the existence of defects in the material and on the surface of the glass structures. Therefore, the production process needs to be optimized in order to prevent such defects. Furthermore, an additional protection against outer influence like air humidity or the physical contact to other materials is required.
Experimental investigation and CFD simulation of organic peroxide pool fires (TBPB and TBPEH)
(2010)
Time averaged mass burning rate (m˙′′f ), flame length (H), temperature (T ), irradi- ance (E) and surface emissive power (SEP ) of TBPB (tert -butyl peroxybenzoate) and TBPEH (tert-butyl peroxy-2-ethylhexanoate) pool fires are measured for six pool di- ameters (d = 0.059 m, 0.107 m, 0.18 m, 0.5 m, 1 m and 3.4 m) at BAM in house and outside test facility. The measured heats of combustion (–Δhc) of TBPB and TBPEH are 30113 kJ/kg and 34455 kJ/kg and the specific heat capacities at constant pressure (cp) are 1.8 kJ/(kg K) and 2.1 kJ/(kg K) respectively. The measured m˙′′f of TBPB and TBPEH pool fires are in the range of 0.37 kg/(m2 s)≤ m˙ ′′ f ≤ 0.83 kg/(m2 s) and show little dependence on the pool diameter d, and are four to sixty times higher (for d = 1 m) than that of hydrocarbon pool fires. It is shown that the mass burning rates of the investigated organic peroxides can be represented as an exponential function of the self-accelerating decomposition temperature (SADT). Low SADT implies that the organic peroxide pool fires burn at a much higher m˙′′f than hydrocarbon pool fires. Fuel Froude numbers (Frf) of TBPB and TBPEH are 5 to 100 times (depending on d) higher than for hydrocarbon pool fires. Due to higher Frf the H of TBPB and TBPEH (measured with a S-VHS Videocamera) are found to be two times larger (d = 1 m) than corresponding pool fires of hydrocarbons. Heskestads flame length correlation predicts the Hd (d = 3.4 m) of TBPB and TBPEH pool fires much better than Thomas and Fay correlations. The measured time averaged flame temperatures T (d = 3.4 m) for TBPB and TBPEH pool fires are in the range of 1400 K ≤ T ≤ 1500 K and are 200 K to 300 K higher than for JP-4, kerosene and gasoline. The irradiances of the TBPB and TBPEH pool fires measured by radiometers are E (Δy/d = 0.3) = 45 kW/m2 and E = 98 kW/m2 which are two to ten times higher in comparison to the corresponding n-pentane, super gasoline and diesel pool fires. So the thermal safety distances for organic peroxide pool fires are larger by a factor four in comparison to the hydrocarbon pool fires. An infrared thermography system is used for the determination of SEP of TBPB and TBPEH pool fires. The values of surface emissive power for TBPB and TBPEH are SEP (d = 3.4 m) = 196 kW/m2 and SEP = 258 kW/m2 and thus the SEP are by a factor of approximately two higher than for hydrocarbon pool fires. A self-sustained pulsating Hd (’W’-Effect) is found in TBPB pool flames and is further analysed to explain the reason of occurance on the basis of chemical structure of the fuel and discontinuous heat flux back from flame to the liquid pool. CFD simulations of TBPB and TBPEH pool fires at d = 0.18 m, 0.5 m, 1 m, 3.4 m and 8 m are carried out using the Unsteady Reynolds Averaged Navier Stokes (URANS) equa- tions. The three-dimensional geometries have been discritized with unstructured hybrid grids, with the number of cells in the range of 1 million. Depending on the grid resolu- tion and the pool diameter time steps of 0.0001 s ≤ Δt ≤ 0.01 s for the CFD simulations are used. For solving the discritized equations a finite volume based implicit solver AN- SYS CFX has been used. For modelling the combustion, stoichiometric combustion for both peroxides are assumed. The temperature dependence of the reaction rate has been determined by the Arrhenius approach. For modelling the combustion eddy dissipation concept (EDC) model has been used. For turbulence buoyancy modified k- � and SAS (Scale Adaptive Simulation) turbulence models are used. For the thermal radiation and soot mass fraction discrete transfer radiation model and Magnusson soot model have been used. A new method is suggested for the prediction of mass burning rate (m˙′′f ) by CFD simula- tion. Both peroxide pool fires show approximately constant mass burning rate indepen- dent of d whereas m˙′′f of TBPEH are under predicted at the beginning but show relatively good agreement with measurements for large pool diameters (d = 1 m). In case of TBPB the CFD simulation over predicts the mass burning rate m˙′′f of small TBPB pool fires and shows a continuous decrease with d. CFD predicts the flame length H close to the measured data provided that the constants in Thomas equation are modified. The CFD predicted time averaged surface emission flame temperatures of TBPB and TBPEH pool fires (d = 3.4 m, 1437 K and 1542 K) are in good agreement with the measured time averaged flame temperatures. The CFD predicted SEP for TBPB and TBPEH pool fires (d = 3.4 m, 217 kW/m2 and 288 kW/m2) are also in agreement with the measured values. From the CFD predicted irradiance ECFD it is possible to determine the thermal safety distances from large pool fires of hydrocarbons and organic peroxides.