•Effect of particles agglomeration in the γ′ free zone on the damage evolution for a single-crystal Ni-based Superalloy is discussed.•A new framework based on crystal plasticity and XFEM is well ...modelled to capture multiple micro-cracks.•Particle fracture and decohesion strengths were determined for both AlN and Al2O3 particles in the γ′ free region.•Interfacial decohesion process, particle fracture, and matrix cracking was together modelled.
Non-metallic inclusions are known to be a significant cause of microstructural deterioration in the single-crystal Ni-based superalloy DD6, resulting in the initiation and propagation of damage. Previous experimental results have revealed that AlN particles are dispersed individually in several regions within the γ′ free zone, situated beneath the oxide layer. However, areas where these particles agglomerate, whether they contain solely AlN particles or clustered with Al2O3 particles, are the most damaging and attractive sites for surface cracks arising in the oxide layers. Interfacial debonding between the particle and matrix, particle fracture, and matrix cracking caused damage initiation in the agglomerate region. To study the damage behavior of the agglomerate area belonging to the γ′-free region and how those particles influence the mechanical drivers of damage nucleation, we developed a new framework based on the Extended Finite Element Method (XFEM) and crystal plasticity (CP) theory. We employed a creep-damage model in a crystal plasticity (CP) framework combined with the XFEM approach to predict the micro-cracks in the agglomerate area (γ′-free zone). We also used the cohesive zone method (CZM) to simulate the interfacial debonding between the γ′ free region and each of the AlN and Al2O3 particles, and cohesive behavior-based XFEM is used to model the particle fracture. The fracture properties of the interface layer between the γ′ free zone and particles, as well as the agglomerated particles, were determined., followed by an assessment of the effects of particle agglomerate on damage mechanisms, including interfacial debonding, particle fracture, and matrix cracking in the agglomeration region. Interfacial partial decohesion of agglomerated AlN particles was prevalent with interparticle microcracking, making the agglomeration area between the AlN particles attractive for surface cracks. In contrast, the damage initiation delay with relatively low rates caused by particle fracture was the damage behavior of the agglomerated Al2O3 model. The consistency between numerical and experimental results confirmed the ability of this framework to capture microstructure-sensitive microcracks.
The aim of the present study was to investigate how variations in the microstructure of 316L austenitic stainless steel influence the tool wear during machining. A detailed comparison between two ...workpieces of 316L, supplied by different producers was made regarding their microstructures and the resulting tool wear during machining. Machining the two workpieces resulted in distinctively different tool wear responses. During the tool life tests, machining one of the workpieces resulted in a steady increase in tool flank wear and the criterion of maximum flank wear land was reached after about 15 and 5 min at low and high cutting speed respectively. However, no significant flank wear was observed when machining the other workpiece under the same cutting conditions. Post-test characterization of the worn tool surfaces showed that tool wear by dissolution/diffusion were the main wear mechanisms for cutting both workpieces. The distinct differences in tool wear progression were linked to varying micro-constituents present in the two workpieces. Specifically, the main factor controlling the wear was attributed to differences in the composition of oxide inclusions. In the one workpiece's case, the specific composition and hence mechanical properties of the inclusions gave rise to a stable protective layer covering the tool surface. During cutting, this inclusion layer was acting as a diffusion barrier, thereby suppressing tool wear by dissolution and hence limiting the progression of flank wear. In contrast, the characteristics of the oxide inclusions present in the other workpiece did not favor the formation of a protective layer on the cutting tool surfaces.
•Varying tool wear rates can occur when cutting same steel from different suppliers.•Cutting forces, grain size and hardness of workpieces would not explain tool wear.•Inclusions can form protective layers on tool surfaces, lowering the wear rates.•Protective layers decrease overall tool wear mainly by suppressing dissolution.
The character, morphology and distribution of inclusions has been studied in additively manufactured 17-4PH stainless steel parts produced from gas-atomized powder by selective laser melting. A ...combination of advanced electron microscopy techniques has been used to show that such parts contain oxide inclusions ranging from a few nm to tens of μm across. The coarser inclusions have morphologies that mimic the oxides which accumulate between melt tracks at the build surface, suggesting that these inclusions are incorporated into the build during deposition of subsequent layers. Such features could have deleterious anisotropic effects upon the fatigue and fracture resistance of the built parts. The finer inclusions are equi-axed, reside preferentially at grain boundaries, and could inhibit grain growth during processing via Zener pinning effects. The chemistries of the inclusions include elements such as Al that are not part of the alloy specification. This indicates that the inclusions are exogenous defects that are entrained in the melt during gas atomization due to interactions with the crucible or nozzle materials. Examples of encapsulated oxide material in the powder feedstock support this explanation. These observations highlight the need for careful control of powder pedigree when using additive manufacturing for critical structural components.
Display omitted
•Inclusions consisting of amorphous oxides of silicon and aluminum are observed in additively manufactured 17-4 PH samples.•The largest inclusions are many microns across with characteristic shapes, and contain crystalline alloy nanoparticles.•Oxide phases with similar morphologies are found between the tracks at the build surface of the samples.•The origin of the oxides is shown to be entrained refractory material within the powder feedstock.•These observations highlight the importance of powder pedigree in metal additive manufacturing.
•In the current paper, newly sintered ceramic substrates with various compositions (Al2O3∙MgO, CA 6, CaZrO3, AZT (Al2O3 based substrates containing ZrO2 and TiO2)) were tested in contact with liquid ...Armco iron under a low oxygen partial pressure in the heating microscope.•The most of the tested substrate showed no reaction with molten iron, except for AZT substrate. The formed reaction layer containing Fe, Si and Mn which were transferred from iron side.•Non-metallic inclusions before and after the experiments were analysed using an automatic scanning electron microscope combined with an energy dispersive X-Ray spectroscope. The total number of the inclusions in the iron samples after all interaction experiments greatly decreased. The deposition and agglomeration of inclusions, and the evaporation of some elements were the factors for the inclusions decreasing.•The stability of oxides consisted of substrates under low oxygen partial pressure was confirmed by the thermodynamic calculation.
Ceramic substrates (Al2O3·MgO, CA6, CaZrO3, AZT (Al2O3 based substrates containing ZrO2 and TiO2)) were interacted with liquid Armco iron under a low oxygen partial pressure in the heating microscope. No noticeable dissolution of the ceramic substrates and reaction was observed for most of the substrates. Furthermore, deposition of SiO2 and Fe was found on the cross section between iron and these substrates. However, for the substrate AZT, a reaction layer of complex composition was found on the interface between the iron sample and the AZT substrate. After all experiments, the total number of inclusions in the iron samples greatly decreased. The deposition, agglomeration of inclusions, and the evaporation of some elements were the factors for the inclusions decreasing. Furthermore, the stability of oxides consisted of substrates under low oxygen partial pressure was confirmed by the thermodynamic calculation.
Bearing failure is a cause of concern in a variety of machinery such as turbines, transmissions, drills, engines, etc. It is often associated with rolling contact fatigue (RCF) triggered from damage ...initiation at non-metallic inclusions (NMI’s). Experimental evidence shows that damage initiation lifetime is highly sensitive to the NMI characteristics and its bonding with the steel matrix. This study numerically investigates the role of NMI features and its bonding with the steel matrix on damage initiation lifetime. NMI characteristics modelled in this study are derived from an experimental investigation of a failed bearing. Simulation results highlight a near to instantaneous debonding at the matrix-inclusion interface followed by accelerated crack initiation. The critical depth for damage initiation shifts towards the surface with the increase in friction coefficient between roller and raceway. The simulations also reveal that larger inclusions show earlier damage initiation, indicating a size effect. The damage hotspots from the simulation results were compared with experimental findings and a hypothesis for crack initiation from a NMI is put forward.
Display omitted
•Near to instantaneous debonding occurs around the matrix-inclusion interface.•Presence of debonding increases the fatigue damage.•The critical depth for damage initiation shifts towards the surface with the increase in friction coefficient.•The simulations also reveal that larger inclusions show earlier damage initiation, indicating a size effect.•Pragmatic approach to estimate fatigue initiation lifetime based on material hardness.
In this study, to investigate the interfacial reactions among high-Mn high-Al steel, MgO-C refractory, and refining slag, a series of lab-scale experiments and thermodynamic calculations are ...performed. For the steel-refractory interface, we observe an interface layer composed of MgO·Al2O3 and a transition layer composed of a CaO-Al2O3 phase, along with trace MgO·Al2O3 and MgO particles. Following the interfacial reaction between the refractory and refining slag, an interface layer and an infiltration layer are detected. Subsequently, the cleanliness of high-Mn high-Al steel is investigated based on the multi-interfacial reactions. The results indicate that the evolution process of oxide inclusion follows the reaction Al2O3→MgO·Al2O3→CaO-MgO-Al2O3, with the addition of refining slag and MgO-C refractory. Additionally, a trend of aggregated AlN inclusion is observed. In certain cases, aggregated AlN inclusions are attached to the MgO·Al2O3 or CaO-MgO-Al2O3 inclusions.
The modification of MgO·Al2O3 spinel inclusions into less detrimental mixture phases of CaO–MgO–Al2O3 plays an essential role in refining calcium-treated aluminium killed steels. This study uses ...Raman spectroscopy for the characterisation of binary phase samples that contain MgO·Al2O3 spinel and calcium aluminate (CaO)x–(Al2O3)y phases. Samples were synthesised from MgO·Al2O3 spinel (MA), Al2O3 and calcium aluminate phases to achieve binary samples of CA–MA, C3A–MA, C12A7–MA and Al2O3–MA with varying phase fractions. The study also examined the possibility of a slight variation for non-stoichiometric spinel samples below the 1600°C region in an MgO–Al2O3 binary system. The relative intensities of the Raman band were used for the quantification of the phase fractions. For a quantitative prediction, linear regression calibration models were identified for each of the studied systems. This work demonstrates the use of Raman spectroscopy for the characterisation of calcium aluminate phases of CA, C3A, C12A7 and magnesium aluminate spinel phases along with Al2O3 and its potential application in inclusion characterisation.
The initiation site of pitting corrosion on type 304 stainless steel was quantitatively analysed using an automatic inclusion analyser and electron backscatter diffraction. In a sample anodically ...polarised at 0.43 VSSE, the fraction of stable pits initiated from inclusions was only 23 %. Pits initiated from sites other than inclusions probably initiated from grain boundaries, strain-induced martensite and scratches. In mixed inclusions of MnS and oxides, the probability of pit initiation increased with the S concentration. The Ti concentration had no effect on the probability of pit initiation, suggesting that whether Ti exists as TiO2 or TiN is important.
Display omitted
•The initiation sites of pitting corrosion on type 304 stainless steel were quantitatively analysed.•Most pits initiated from grain boundaries, strain-induced martensite and scratches.•Among inclusions, MnS was the most likely to initiate pits.•In mixed inclusions of MnS and oxides, the probability of pit initiation increased at higher S concentrations.•The elements released from Ti-based inclusions may affect the tendency of pit initiation.
Nickel-based superalloys are used in high strength, high-value applications, such as gas turbine discs in aero engines. In these applications the integrity of the disc is critical and therefore ...understanding crack initiation mechanisms is of high importance. With an increasing trend towards powder metallurgy routes for discs, sometimes unwanted non-metallic inclusions are introduced during manufacture. These inclusions vary in size from ∼10 μm to 200 μm which is comparable to the grain size of the nickel-based superalloys. Cracks often initiate near these inclusions, and the precise size, shape, location and path of these cracks are microstructurally sensitive. In this study, we focus on crack initiation at the microstructural length scale using a controlled three-point bend test, with the inclusion deliberately located within the tensile fibre of the beam. Electron backscatter diffraction (EBSD) is combined with high spatial resolution digital image correlation (HR-DIC) to explore full field plastic strain distributions, together with finite element modelling, to understand the micro-crack nucleation mechanisms. This full field information and controlled sample geometry enable us to systematically test crack nucleation criteria. We find that a combined stored energy and dislocation density provide promising results. These findings potentially facilitate more reliable and accurate lifing prediction tools to be developed and applied to engineering components.
Display omitted
•High resolution digital image correlation.•High resolution electron backscatter diffraction.•Crack nucleation.•Non-metallic inclusion.
The actual service life of wind turbine gearboxes is often well below the desired 20 years. One of the prevalent failure modes in gearbox bearing raceways is white structure flaking (WSF) in as ...little as 6-24 months of operation by the formation of axial cracks and white etching cracks (WECs) with associated microstructural change called white etching areas (WEAs). Despite these failures having been observed for two decades in various industries, the drivers and mechanisms for their formation are still highly contested. Discussed in this review are methods for searching and analysing WECs, mechanisms for WEA microstructural change, WEC initiation and propagation theories, WSF formation drivers and finally technologies and processes offering resistance to WSF. This updated review serves as a recap, comprehensive update on findings, current focus areas and remaining challenges.
This paper is part of a Themed Issue on Recent developments in bearing steels.
PDF
