The effect of soft impingement is modelled for the isothermal, diffusion-controlled growth of spherical precipitates in the solid-state. A two-stage soft impingement phase, isotropic growth in 3D and site saturation are assumed, which lead to a mean field approach for soft impingement. A mass balance is used in combination with a non-linear concentration profile approximation, instead of the ordinary linear concentration profile, to improve the description of growth under conditions of overlapping diffusion fields and low supersaturation. In addition, the capillary effects, dislocation assisted growth via pipe diffusion, and boundary assisted growth via boundary diffusion are discussed. The model has been applied to the growth of TiC-precipitates during tempering of a quenched Fe-C-Mn-Ti steel. The results are in good agreement with published experimental data of precipitate growth.

Advancing our understanding of the mechanisms of phase transformations is an efficient pathway to designing new steels with enhanced, tailor-made mechanical properties at minimal experimental cost. This dissertation investigates the bainitic transformation in steels. The mechanism of bainite formation is still not well understood, as there is no consensus on the role of carbon diffusion during the growth of bainite. Thus, it is difficult to understand and predict chemical and microstructural effects on the kinetics of bainite formation, such as the effect of prior austenite grain size and the presence of prior martensite. A better understanding of such effects may allow the design of new high performance bainitic steels with lean chemical composition and energy efficient heat treatments. In this dissertation, a new analytical model of the kinetics of bainite formation based on the displacive-diffusionless theory is proposed. This new model can reproduce and offer insights into the effect of prior austenite grain size and prior martensite on the kinetics of bainite formation. Following the understanding achieved with the model, the accelerating effect that martensite has on bainite formation kinetics is used to develop novel advanced high strength steels for the automotive industry that can be manufactured in existing continuous annealing lines thanks to fast bainite formation. Finally, to better understand the nucleation and growth of bainite, a real time observation of bainite formation using in situ transmission electron microscopy (TEM) is shown and discussed.

Continuous casting is the dominant manufacturing method of steels, accounting for over 90% of global steel production. Mould slags, which are calcium silicate glasses, play a crucial role in this process due to their multifunctional purpose, directly impacting the safety, quality, and efficiency of steel production. As the steel industry transitions toward greener practices, new challenges arise, including the need for fluorine-free mould slags to meet increasingly stringent safety and environmental regulations. Historically, mould powder development has relied heavily on trial-and-error approaches, limiting the development rate of these materials for new demands. This thesis is part of a broader research effort aimed at understanding the structural transformations that drive the desired material properties of mould slags. In this study, various characterization techniques—such as X-ray diffraction (XRD), scanning electron microscopy (SEM), and high-temperature Raman spectroscopy (RS)—were employed to investigate the structure of an industrial mould slag, with a focus on its heat transfer properties. Samples were melted and then solidified followed by three different isothermal holding temperatures to replicate the thermal cycling experienced during casting. The multi-method characterization revealed that the industrial mould slag (S1) consists of three distinct phases: cuspidine, combeite, and a glassy matrix. These phases were present regardless of the isothermal holding temperature. The microstructure after solidification featured large faceted cuspidine crystals embedded within a mixture of smaller dendritic combeite crystals and the glassy phase. High-temperature RS measurements showed that the formation kinetics of the S1 mould slag were faster than anticipated based on previous research. This highlights the necessity of multi-method characterization, and not to rely only on the chemical composition to fully understand mould slags and adapt them to future industry requirements. While there is room for improving the accuracy of the utilized technique, high temperature RS measurements have demonstrated their value in mould slag research and, with further refinement, hold the potential to unlock new research opportunities.

VDL Weweler has developed a novel heat treatment approach for their trailing arms, which eliminates the traditional austenisation step and instead capitalises on the inherent heat used for hot-forming to austenise the steel before quenching. This not only saves energy, but surprisingly also improves the lifetime of the products. The underlying reasons for this improvement in fatigue life are the central focus of the present study. Two trailing arms were used for this study, one from the traditional ‘Reference’ process, and one from the new ‘Gasloos’ process. Both were produced from the same heat of raw 51CrV4 steel bars that are inductively heated to 1200°C before hot-forming, after which Gasloos is immediately quenched to 180°C. Reference is left to cool to 600°C before being reheated in a gas oven at 900°C for one hour and is finally also quenched to 180°C. After quenching, both are tempered at 450°C for one hour, quenched to room temperature, and shot peened. Reference also underwent the usual painting procedure, Gasloos was collected before this. Both underwent fatigue bending tests, where the Gasloos trailing arm lasted about seven times longer. Fracture surfaces and unfatigued parts were collected from both trailing arms for analysis. Since it was obvious early on that the Reference trailing arm has a partially intergranular fracture, a literature survey was performed to assess the possibility that grain boundary segregation and embrittlement would occur. Scanning Electron Microscopy (SEM) combined with Energy Dispersive X-Ray Spectroscopy (EDS) was used to perform fracture and microstructural analysis, as well as to measure grain boundary segregation. X-Ray Diffraction (XRD) analysis was used to measure the depth profiles of the residual stress to assess the effectiveness of the shot-peening. Samples for XRD were cut using Electronic Discharge Machining (EDM). In both trailing arms, cracks initiated on slag inclusions present from the steelmaking process. Reference initiated on an inclusion 10μm in size, while Gasloos initiated on an inclusion 45μm size, 4.5 times bigger. Slag inclusions are on average 16.09μm in size. The Reference fracture surface is 16% intergranular, while Gasloos is fully transgranular. Carbide density on the PAGB is more than twice as high in Reference, while overall carbide density is 10% higher in Gasloos. Solute segregation to the PAGB could not be
measured with EDS. XRD analysis shows that the compressive residual stress from shot-peening is 1.5 times higher and twice as deep in Gasloos compared to Reference. This is due to the oxide scales, containing 17.49 weight% oxygen, that comprise the surface of Reference, making it very brittle and filled with voids. Gasloos does not have these oxide scales at the surface.

In order to scale the planar flow casting (PFC) process to industrial levels a Python model is created. The model which is based on a combination of empirical and theoretical equations, is used to explore the limits of the process. The model is independently verified with a Comsol simulation, which is extended with solidification and heat transfer physics. The model is then used as a bases for a system and control simulation of the complete planar flow casting system. Extra elements are added, such as thermal expansion, out of roundness of the wheel, crucible and actuator mechanics, and sensor behaviour. The controllability of the system using a PID controller is presented. With this knowledge, the design requirements for the adjustment of the existing PFC machine were defined, qualitatively as well as quantitatively, for the nozzle, pressure system and gap sensor. Following this, the microstructure of the material produced by the free flow casting (FFC) process is analysed. The microstructure is measured using XRD, SEM and EDS techniques. These measurements are then compared with a model based on the Scheil equation and a kinetic phase diagram, developed for the aluminium silicon alloy. The microstructure found consisted of primary silicon particles in which a significant amount of aluminium was dissolved, as predicted by the material model. The silicon particles are dispersed within an aluminium silicon eutectic matrix with rounded needle like features. The silicon particle size is in the order of 5 μm, whilst the feature size in the eutectic is in the order of 100 nm.

Inelastic deformation is a common but often neglected phenomenon in experimental analysis of metal deformation and in contacts. This neglect leads to degraded measurement accuracy of material properties. Therefore a need arises for material models that a priori incorporate inelasticity. These material models must be simple and comprehensive to have the highest impact in society. This thesis addresses three main sources of inelasticity, namely anelasticity and plasticity in metals and viscosity in contacts. Inelasticity is a dissipative mode of deformation that is mechanically recoverable for anelasticity and viscosity, and irrecoverable for plasticity. We connect the fundamental properties and structures of metallic and soft matter constituents with experimentally accessible measures. The presented models will aid in the development of materials with specific properties that meet the needs of industry.

Chapter 2 presents an analytical model of the tensile test tangent moduli and yield points for single-crystallite metals with spatially uniform and nonuniform dislocation distributions across slip systems. The moduli and the onset of plastic flow show a notable dependence on initial dislocation character, spatial dislocation distribution, and loading direction with respect to crystallographic orientations. An improved methodology accounts for elastic compressibility and anisotropy, and the geometric structure of crystal lattices when one measures dislocation network geometry in single metallic crystallites.

Chapter 3 contains a seamless, unified stress-strain treatment of dislocation-driven deformation. This treatment combines the three deformation mechanisms of elastic bond stretching, stable dislocation glide, and unstable dislocation glide. The model’s yield criterion connects the bowing out of local dislocation links and global dislocation multiplication. A semi-empirical relation is constructed for the evolution of the dislocation network structure with uniaxial loading.

Chapter 4 formulates a macromechanical model of the yield point phenomenon under invariant plane conditions. The heterogeneous stress state across the Lüders front and the plastic flow inside the Lüders band are accounted for. The Lüders band orientation with respect to the tensile direction is not unique; the orientation changes with material properties and tensile specimen geometry by the stress concentration at the front. The model serves to approximate constitutive parameters independent of the test conditions.

Chapter 5 elucidates the interplay between adhesion and roughness by modelling the retraction of rigid, wavy indenters from viscoelastic substrates. Viscoelasticity governs adhesive hysteresis across all loading rates, and even in the presence of roughness-induced mechanical instabilities. This confirms the central role that viscoelasticity must play in experimental measurements in the presence of adhesive interfaces in soft matter contacts.

Chapter 6 examines the static, quasi-static, and dynamic trajectories of a base-excited mass-spring-damper system in the presence of friction. The differences between the dynamic and the quasi-static solution in engineering problems with viscous, static, and dry friction are assessed. The omission of inertial contributions will under-predict dissipation at both low and high excitation frequencies. This chapter is a guide for future (multi-scale) numerical modelling efforts on adhesion and interface friction, and the hysteretic deformation of metals.

Chapter 7 is a general discussion on the impact of inelasticity in metals, that follows from Chapters 2 and 3, the measurement of the yield point phenomenon in Chapter 4, and numerical modelling of dissipative contacts in Chapters 5 and 6. The four models as presented in Chapters 2-6 are readily applicable in experimental measurements and future numerical models. The importance of accounting for inelasticity in experimental measurement and modelling of the yield strength in metals, and adhesive dissipation in soft matter contacts is emphasised. Finally, the state of the art in research on the three main sources of inelasticity and potential applications of the presented models are enumerated, which serve as starting points of future research.

51CrV4 is a spring steel commonly used for the production of trailing arms in the suspension assembly of trucks. It is essential for these products to be predictable and reliable. The presence of carbide banding in 51CrV4 steel can be detrimental to this predictable and reliable behavior. Therefore, in this research, the formation of carbide-banded microstructure in 51CrV4 steel will be investigated. The bending stresses undergone by trailing arms make it crucial to understand the relationship between surface depth and carbide band formation. Therefore, the goal of this research is to understand the relation between the surface depth and band formation in 51CrV4 steel and how banding differs among 51CrV4 steel from different manufacturers. This was done using a variety of testing methods which include microscopy, hardness, XRD, SEM, EDS and EPMA. These testing methods were used to understand the formation of carbide bands in 51CrV4 as well as the underlying physical properties that lead to the carbide band formation. Overall, a significant difference could be seen in the formation of carbide bands across the steel from different manufacturers. These differences in carbide banding properties could be clearly linked to the difference in the physical properties of the steels. The depth at which carbide band formation tends to start could be most clearly related to the micro-segregation peak concentration. Furthermore, the manner in which the carbide bands are built up along the surface depth could be most clearly related to the chromium concentration in the carbide banded regions. Finally, it was found that chromium is expected to be the primary carbide former in this steel as the formation of carbide bands is closely and consistently related to an increased presence of chromium.

The Static Recrystallization (SRX) phenomenon of austenite in C-Mn steels have a large impact on the final mechanical properties of the steel. However, the austenite recrystallization kinetics are difficult to investigate experimentally, due to solid-state phase transformations during cooling to room temperature. The best model alloys to examine SRX are those austenitic alloys such as, Ni-30Fe alloy, that do not experience undesirable phase transformation and that have similar stacking fault energy (SFE) as austenite in C-Mn steels. The hot-working of the Ni-30Fe alloy has been conducted at high deformation temperatures, followed by an uniaxial compression test at a low strain condition (ε = 0.2) with a strain rate of 1s^-1, and an annealing temperature of 900°C (above recrystallization start temperature). Thereafter, the alloy is experimentally investigated by Electron Backscatter Diffraction (EBSD), to characterize the austenite microstructure and to measure the crystallographic orientations. The experimental measurements from the EBSD technique are subsequently analyzed using MTEX software to study the SRX phenomenon that occurs during hot rolling of steels.The technique to separate recrystallized and deformed grains in this research has adopted the Kernal Average Misorientation (KAM) and Grain Average Misorientation (GAM) technique, rather than the previously utilized Grain Oriented Spread (GOS) technique which incidentally produced slower and inaccurate recrystallization kinetics. The initial and final grain size calculations have been conducted by consideration of the annealing twins, and the reported results are in close agreement with earlier conducted work of Sellars model. The EBSD scan which captures the Geometrically Necessary Dislocations (GND) have been experimentally measured after deformation to be approximately 4.5x10¹⁴m^-2, by utilizing the approach of KAM, curvature tensor and the Burgers vector with an estimated 13% error. Additionally, the effect of recovery has been captured with the continuous decrease in stored energy as a function of time, where the GND density is estimated to be 7.3x10¹²m^-2 after the annealing treatment.The GND density present in the deformed microstructure is significant to provide the net driving force necessary for nucleation, based on the validated strain-induced-boundary-migration (SIBM) mechanism. The initial nucleation site is observed at triple junctions, and the selected nucleation criterion factor for the captured GND densities, estimates the driving force for nucleation to favour single sub-grain SIBM. The minimum critical sub-grain size required for bulging is calculated to be 3.40 ± 0.44 μm after deformation, which is an increasing parameter with time, as a result of the recovery process. The number of recrystallized grains captured at different time intervals using the validated GAM technique assists in experimental nucleation rate calculations, that have been subsequently converted for a 3D microstructure, in order to compare with the nucleation framework of Rehman's model. The recrystallization nucleation process is experimentally observed to obtain peak values at 2s and continues until all the GNDs are annealed out of the specimen. There is a large discrepancy in the nucleation rate predicted by the model, which is much higher with respect to the experimental observations reported. Finally, the experimental recrystallization kinetics is observed to be in close agreement with the JMAK model during the initial phase of the annealing treatment, but deviates in the later phase of the recrystallization process.

Bainitic steels are the type of steels which are widely in use, especially in automotive industries, for its good mechanical behavior, i.e., with coexistence of strength, ductility and fatigue resistance. A better understanding on the mechanism and transformation kinetics of bainite are necessary in order to improve the performance of the bainitic steel further. This research mainly studies the kinetics of bainitic transformation during isothermal treatments. The chemical composition, in particular silicon (Si), affects the transformation kinetics of bainite, and therefore the fractions of retained austenite and bainitic ferrite. The current research focuses on the bainite transformation kinetics during isothermal heat treatment. Steel specimens are austempered at 250°C, 300°C and 350°C and held for 30, 60 and120 minutes, respectively. The steel in the current research has the alloy composition of Fe-0.61C-1.62Si-0.85Mn-0.32Cr (wt.%). Microstructures observed using optical micrographs consist of bainite and retained austenite after the austempering process. Quantitative measurements of the retained austenite (RA) fraction are performed by magnetization technique. The results show that, at austempering temperatures of 250°C and 300°C, the fraction of retained austenite decreases gradually with increasing holding times and increases with increasing austempering temperatures. However, a different affect is observed in the steel austempered at 350°C for 30minutes. The fraction of retained austenite increases from 30 to 60 minutes and subsequently decreases from 60 to 120 minutes. In order to study the effect of retained austenite on hardness resulting in decrease in overall hardness with increasing austempering temperatures and increases with increasing hold duration. The JMAK model has been fitted to the observed fraction of austenite as function of time and temperature, which results in the parameters such as the rate constant and Avrami exponent. The fitted results suggest a one-dimensional grain growth of the bainite in this steel. The predictions from a thermodynamic analysis using the para-equilibrium model and the T0-temperature are compared to the experimentally observed fractions of RA, which results in obtaining higher fractions than obtained experimentally. This thermodynamic analysis predicts an increase in the fraction of remaining austenite with increasing austempering temperature as observed experimentally in most cases. The thermodynamic study on the effect of Si concentration (1.62 wt.%)results shows that there is insufficient Si present in the steel to hinder the formation of cementite.

In this study, non-destructive methods are applied to quantify the influence of concrete degradation by geologic conditions on the ability to retain nuclear waste deep underground in The Netherlands. Nuclear reactors play an important role in all four scenarios of the recent report of the IPCC to reach the CO2 reduction goals. Safety related criticism against nuclear energy technology are less relevant for Generation IV nuclear power plants, of which several are scheduled to be constructed in Europe within the next few years. This is partially due to the inherent safety features of Gen IV power plant designs, where disasters experienced before (e.g., Chernobyl, Fukushima) are technically impossible. Concrete samples provided by COVRA are degraded by immersion in a solution that contains chlorides and sulphates in concentration levels that are similar to geologic conditions. The influence of concrete degradation on the internal structure of concrete has been investigated using X-ray and Neutron radiography. Pore size distributions, pore geometry and sorptivity are presented. In order to validate the results, standardized tests are performed and the results are compared.

Hot-dip galvanizing is a well-known process to increase the corrosion resistance of steel. As a by-product dross is formed in the Zn-bath. The dross particles are composed of Fe, Al and Zn in the form of Fe2Al5Znx and interact with the hardware that is submerged in the Zn-bath and eventually accumulate on the surface of the hardware. This accumulation of dross on the hardware is known as dross build-up.

Dross is formed in the Zn-bath as a result of the dissolution of Fe from the steel strip. This Fe reacts with Al present in the liquid Zn forming the Fe2Al5Znx dross particles. Once the hardware is submerged in the liquid Zn a thin compact Fe2Al5-layer is formed on top of the surface of the hardware, known as the diffusion layer. This diffusion layer acts as a barrier for Fe towards the Zn-bath and Zn towards the hardware surface. Once the diffusion layer is formed dross particles precipitate on top of this diffusion layer followed by a slow growth of the intermetallic dross particles. The diffusion layer and accumulated intermetallic dross particles are known as the dross build-up. The thickness of this dross build-up depends on immersion time, bath temperature and bath chemistry and varies typically between 90 μm up to several millimetres.
The dross build-up is thought to be a critical factor in the bath hardware lifetime. The surface quality of the bath hardware directly influences the quality of products produced on the galvanizing line. More dross in the Zn-bath could lead to more defects in the coating, which could lead to failure when the steel strips are processed further, or giving the coating on the steel strip a bad appearance. Because of the dross build-up on the hardware, the bath hardware is changed every 4-6 weeks. By controlling the dross in the Zn-bath, the service lifetime of the hardware could possibly be extended. When the service lifetime of the hardware is extended the maintenance downtime and costs of the galvanizing lines are reduced.

Recently, Tata Steel introduced a new type of Zn-coating: MagiZinc (MZ). This type of Zn-coating differs from the conventional Zn-coating: Conventional Zn-coating (GI) consists of Zn with 0,30 wt% Al whereas MagiZinc consists of Zn with 1,60 wt% Al and 1,60 wt% Mg. This difference in composition has an influence on the dross build-up formed on the hardware. At Tata Steel, there is some evidence that the dross build-up layer on the bath hardware created in the conventional Zn-bath is diminishing when the hardware is submerged in the MagiZinc-bath.
This thesis project aims to identify the characteristics of this cleaning behaviour in MagiZinc of the dross build-up that is formed on bath hardware when submerged in conventional Zn.

Based on the results obtained from experiments in this study it can be concluded that the delamination process of the intermetallic dross particles is a combination of intergranular diffusion of Zn and crack formation as a result of thermal shock.

By changing the baths from conventional Zn to MagiZinc the composition of the bath changes. As a result the thermodynamic stability of the intermetallic Fe2Al5Znx dross particles change with respect to the liquid Zn-phase in such a way that the intermetallic dross particles partly dissolve in the liquid MagiZinc. As a result, the intermetallic dross particle/liquid Zn interface changes from a faceted to a curved interface. As a consequence of this change in structure intergranular diffusion can take place between the intermetallic dross particles. By the intergranular diffusion of Zn the cohesion of the grain boundaries of the intermetallic dross particles is reduced. This reduction in cohesion is probably the start of the delamination of the intermetallic dross particles, breaking into smaller pieces at the grain boundaries when enriched with Zn.
The diffusion layer remains largely unaffected in the delamination process due to the better adhesion to the hardware surface compared with the adhesion of the intermetallic dross particles to the diffusion layer. The better adhesion of the diffusion layer to the hardware surface is the result of diffusion of Cr and Ni from the 316L SS substrate into the liquid Zn at time of immersion. The area from where Cr and Ni are dissolved, Al is diffusing into the 316L SS substrate forming the diffusion layer. Simultaneously the intermetallic dross particles form at the hardware surface. This layer is mainly formed from Fe and Al from the bath, not from the 316L SS substrate.
Due to this limited bonding between the intermetallic dross particles and the diffusion layer the intermetallic dross particles are more prone to crack formation due to thermal shock.

By changing the Zn-bath from conventional Zn to MagiZinc and vice versa the hardware and thus also the dross build-up rapidly cools down to room temperature. This thermal shock creates stresses within the intermetallic dross particle-layer. Due to a difference in thermal expansion between the 316L stainless steel matrix and the intermetallic dross particle-layer large cracks form in the intermetallic dross particle-layer. When the hardware is immersed again in liquid Zn the cracks still exist and these cracks accelerate the delamination process of the dross build-up structure. This observed mechanism is independent from the type of Zn-bath that the hardware is immersed in. The process also takes place when the hardware is taken out a conventional Zn-bath and placed back into a conventional Zn-bath.

The Dual Phase Steels (DP Steels) have attracted interest of steel manufacturers, especially automotive industry, because of a good combination of high strength and ductility. The final DP steel microstructures are formed during continuous heating to intercritical temperatures, then isothermal holding and quenching to form a mix of ferrite and martensite. Final martensite content, morphology and spatial distribution depends on the austenite formed during the isothermal holding. The austenite formation is affected by the progress of ferrite recrystallization during continuous heating. Therefore, it is important to understand the process of ferrite recrystallization.
This thesis aims to study the effect of Si and Mn, which are common alloying elements in steels, on the ferrite recrystallization. The present study extensively uses in-situ 2D X-ray Diffraction (XRD) to investigate the ferrite recrystallization kinetics. This technique is faster and records more data than the other conventional methods used to study the kinetics of recrystallization until now. This study introduces a method to quantify the recrystallized grains using Scanning Electron Microscope (SEM) and Backscattered Electron Detector (BED). This method results in faster analysis of microstructure. The microstructure analysis helped in quantifying the grain growth, its dimensions and type of nucleation. Afterwards, those have been used as input parameters for modified non-isothermal JMAK model. This model is used to obtain the kinetic parameters from the experimental in-situ 2D XRD results by model fitting. The parameters obtained are: rate constant, activation energy and Avrami exponent. In all the cases, the microstructure analysis show site-saturated nucleation and predominantly 2D grain growth. The model fitting reveals that most nucleation cases show site-saturation. The density of nucleation sites is dependent on the pearlite content of the alloys. The boundaries between deformed pearlite and ferrite being the preferred nucleation sites. In some cases, the recrystallization nuclei are also seen at the boundaries between the ferrite grains which correspond to areas with micro-segregation of Mn.
Presence of Mn is leading to an increase recrystallization start temperatures. The solute drag effect of Mn is quite high. Similarly, Si is also retarding the recrystallization. The retardation effect of Si is not as significant as Mn. The solute drag effect seems to be dependent on the velocity of the grain boundaries and consequently, on progress of recrystallization. In presence of Mn, the Si is able to interact with the moving grain boundaries. This points to co-segregation effect between Mn and Si. This effect was found to be strongest for 1:1 Si to Mn atomic fractions, where the lowest growth rates were observed.
KEYWORDS: DP steels, ferrite recrystallization, solute drag effect, in-situ 2D XRD, recrystallization kinetics, SEM, JMAK model, activation energy, co-segregation.