Turbulent Melt Flow Research Articles

A counter-gravity casting system for in situ high-speed synchrotron x-ray imaging characterization.

Counter-gravity casting (CGC) aims to eliminate turbulent melt flow and defect formation during filling and subsequent solidification by pushing high-temperature melt into the mold cavity against gravity with regulated pressure. However, limited by the opaqueness of molten metals and the complexity of the CGC apparatus, it is extremely difficult to directly quantify the high-velocity mold filling and pressurized solidification in real-time. Here, we report the design and characterization of a CGC system capable of in situ monitoring of mold filling and subsequent solidification processes in the synchrotron beamlines by deploying a high-energy, high-speed synchrotron x-ray imaging technique. The high-velocity melt flow and dendrite growth during pressurized solidification have been quantified for systematical process parameter analysis by investigating time-resolved x-ray images of an exemplary Al-Cu alloy. The high-speed imaging results demonstrate that the in situ CGC system provides a useful way to better understand the fundamentals of mold filling, pressurized solidification, and experimental inputs for high-fidelity modeling in scientific and industrial applications.

Dynamics of pore formation and evolution during multi-layer directed energy deposition additive manufacturing via in-situ synchrotron X-ray imaging: A case study on high-entropy Cantor alloy

Blown-powder directed energy deposition (DED) additive manufacturing is impeded for novel alloys processing by perceivable and detrimental porosity. During multi-layer depositions, however, mechanisms of pore formation and evolution remain elusive for developing pore mitigation strategies. Here, conduction-mode multi-layer DED process of an exemplary high-entropy Cantor alloy have been investigated in-situ by high-energy high-speed synchrotron X-ray imaging. Three new pore formation mechanisms are unveiled when depositing first layer and successive layers: gas pore induced by high-velocity powder injection into melt pool, pore generated from swirl shear of turbulent melt flow, and pore trapped by surface wave. Three pore formation mechanisms are reconfirmed: pore inheritance from feedstock powder, pore generation when laser remelting defect-sensitive locations of existing pore from previous layer or unmelted powder attached on the melt pool surface, and pore formation as cooling of melt pool. A unique mechanism for pore elimination is proposed: a counter-Marangoni melt flow is experimentally found in the stable melt pool and contributes to the prolonged pore lifetime at tens of milliseconds scale; pores are prone to coalesce into larger sizes in laser interaction zone and the adjacent location with circulation zone; coalesced larger pores driven by combined effect of Marangoni and buoyant forces easily get eliminated from melt pool. The results of pore formation and evolution dynamics revealed in Cantor alloy provide quantified experimental data for high-fidelity computational modeling and in-depth insights of porosity control for high-entropy alloy printing down to melt pool scale.

Effects of the asymmetric and oscillating turbulent melt flow on the heat transfer and solidification inside the thin slab continuous casting (TSC) mold under the applied electromagnetic brake (EMBr)

The thin slab casting (TSC) is a breakthrough near-net-shape technique for flat products accompanied by rapid casting and solidification rates. The TSC quality hinges on the turbulence, super-heat flow and growth of the solidified shell. The electromagnetic brake (EMBr) is commonly applied to control the fresh melt flow after feeding through a submerged entry nozzle (SEN). Numerical modelling is a perfect tool to investigate the multiphase phenomena in the continuous casting (CC). The presented study considers the heat transfer through the solid shell and water-cooled copper mold including the averaged thermal resistance of the slag skin and the air gap coupled with the turbulent flow and magnetohydrodynamics (MHD) model using an in-house code developed inside the open-source computational fluid dynamics (CFD) package OpenFOAM®. The model is applied to investigate different undesired asymmetric melt flow issues: (i) with the misaligned or (ii) partially blocked SEN; (iii) caused by the mean flow fluctuations with the natural frequencies; (iv) related to the oscillations of the fresh melt jets for the specific SEN designs and casting regimes. The variation of the flow pattern and superheat distribution is studied and presented for different scenarios both with and without applied EMBr.

COMPUTER ANALYSIS OF INFLUENCE OF MELT BLOWING MODES ON LADLE LINING MECHANICAL EROSION

In the process of blowing, turbulent melt flows erode the ladle lining and its thickness de-creases due to mechanical erosion. The thickness of the steel ladle lining gradually becomes thinner with each casting. The degree of erosion depends on the tangential melt speed. In steel production, they try to slow down this wear, because each lining repair costs considerable time and resources. Scientists paid attention to this problem in publications, in particular, on mathematical modeling of lining wear. A large number of conditions of this process are subject to research, in particular, the number and location of blowing tuyeres, as well as blowing power. Firstly, it is necessary to quickly mix the impurity in the melt, and secondly, to preserve the lining of the ladle. Computer visualization and analysis of this process involves its course and results in the form of calculated fields, in particular, wear. The result fields are stored in a database. Also they are added and processed through a specially designed website. It allows researchers to register and fill in the experiment form, as well as add literature sources of data. The list of literature is used in almost all experiments to compare results. Simulation of the process at blowout rates of 40, 60 and 90 l/min and the number of blowout plugs (tuyeres) from one to three showed that the greatest scouring is predicted at the bottom, near the blowout plugs, and the transition to each higher blowout rate increases the scouring intensity by about 15%. Turning off the tuyeres after 1 minute of blowing significantly reduces erosion by at least 35%. If we consider the ladle wall, without disconnecting the tuyere, the flow rate of 90 l/min is the most destructive.

Transient simulation of melt flow, clogging, and clog fragmentation inside SEN during steel continuous casting

Clogging of submerged entry nozzle (SEN) during continuous casting of steel is an undesirable phenomenon leading to different problems like flow blockage, slag entrainment, nonuniform solidification, etc. A transient numerical model for nozzle clogging based on an Eulerian-Lagrangian approach was developed and it covers the main steps of clogging: (a) formation of the first oxide layer by chemical reactions on the steel-refractory interface; (b) motion of non-metallic inclusions (NMIs) due to the turbulent melt flow towards the SEN wall; (c) interactions between the melt, the NMI, and the wall; (d) formation and growth of the clog by the deposition of NMIs on the clog front and the flow-clog interactions; and (e) detachment/fragmentation of a part of clog due to the flow drag force. Clogging in an industrial scale SEN was simulated. The simulated clog front was compared with real as-clogged SENs. The modeling results have successfully explained the SEN clogging induced transient flow phenomenon in the mold region, i.e. the transition from the stable to an unstable and non-symmetrical flow.

Design and optimization of complex single heater for vertical gradient freeze (VGF) grower

A complex single heater is designed for a VGF process to decrease process complexity caused by the use of a multi-heater system. To do this, a set of design criteria is selected to ensure that the final quality of the crystal is acceptable. Then a 3D transient finite element model is coupled with a new optimization procedure. All important physical phenome a such as temperature distribution, turbulent melt flow, thermal stress and crystal front shape are monitored and controlled during crystal growth. Using these, a single heater is designed for 4″ InP VGF crystal growth which meets all design criteria. It is concluded that using the complex single heater, high quality InP crystal growth by VGF process is possible.

Effect of zinc vapor forces on spattering in partial penetration laser welding of zinc-coated steels

A three-dimensional thermal-fluid numerical model considering zinc vapor interaction with the molten pool was developed to study the occurrence of zinc vapor-induced spatter in partial penetration laser overlap welding of zinc-coated steels. The zinc vapor effect was represented by two forces: a jet pressure force acting on the keyhole rear wall as the vapor bursts into the keyhole and a drag force on the upper keyhole wall as the vapor escapes upwards. The numerical model was calibrated by comparing the predicted keyhole shape with the keyhole shape observed by high-speed X-ray imaging and applied for various weld schedules. The study showed that large jet pressure forces induced violent fluctuations of the keyhole rear wall, resulting in an unstable keyhole and turbulent melt flow. A large drag force pushed the melt adjacent to the keyhole surface upward and accelerated the movement of the melt whose velocities reached 1 m/s or even higher, potentially inducing spatter. Increased heat input facilitated the occurrence of large droplets of spatter, which agreed with experimental observations captured by high-speed camera.

Mathematical Modeling of the Early Stage of Clogging of the SEN During Continuous Casting of Ti-ULC Steel

The clogging of the submerged entry nozzle (SEN) during the continuous casting of steel can be divided into two stages: the “early stage,” when the initial layer of the clog covers the SEN refractory surface owing to chemical reactions, and the “late stage,” when the clog layer continues to grow because of the deposition of non-metallic inclusions (NMIs). In this paper, a mathematical formulation is proposed for the build-up of the initial oxide. The chemical reaction mechanism is based on the work of Lee and Kang (Lee et al. in ISIJ Int 58:1257–1266, 2018): a reaction among SEN refractory constituents produces CO gas, which can re-oxidize the steel melt and consequently form an oxide layer on the SEN surface. The proposed formulation was further incorporated as a sub-model in a transient clogging model, which was previously developed by the current authors to track the late stage of clogging. The thermodynamics and kinetics of CO production, depending on the local pressure and temperature, must be considered for the sub-model of early-stage clogging. Test simulations based on a section of an actual industrial SEN were conducted, and it was verified that the clogging phenomenon is related to the SEN refractory, the chemical reaction with the steel melt, the local temperature and pressure, and the transport of NMIs by the turbulent melt flow in the SEN. The model was qualitatively validated through laboratory experiments. The uncertainty of some parameters that govern the reaction kinetics and permeability of the oxide layer is discussed.

Understanding the influence of ultrasonic streaming on solidification of AZ80 magnesium alloy during direct-chill casting using a transient coupling modeling

In this modeling, the influence of ultrasonic streaming on physical fields during direct-chill (DC) casting of AZ80 magnesium alloy is investigated by a transient 2D axisymmetric mathematical model, which couples the ultrasonic field, fluid flow and temperature field using COMSOL Multiphysics software. The influences of input acoustic pressure, inserted depth and shape of ultrasonic horn on physical fields are discussed systematically. Compared with the DC casting, the ultrasonic casting (UC) has more turbulent melt flow, uniform temperature gradient, and shallow liquid sump. The acoustic streaming driving force dramatically changes the melt flow and temperature distribution. The increase of input acoustic pressure from 0.2 MPa to 2.2 MPa enlarges the affected area of ultrasonic streaming. However, a further increase of input acoustic pressure to 4.2 MPa has sight effects on physical fields. With the increase of inserted depth of ultrasonic horn from 10 mm to 50 mm, the circulation caused by ultrasonic streaming is enlarged. The convex surface of the ultrasonic horn promotes the transverse flow of melt compared with the flat horn. Therefore, the transverse temperature gradient and depth of liquid sump are decreased.

Effect of cusp magnetic field on the turbulent melt flow and crystal/melt interface during large-size Czochralski silicon crystal growth

The influences of cusp magnetic field on the turbulent melt flow and the crystal/melt interface are studied by three-dimensional unsteady simulations for 300 mm industrial Czochralski silicon crystal growth. Detailed comparisons are implemented for the melt flow pattern, temperature and velocity oscillations as well as the crystal/melt interface with and without magnetic field. It is found that the melt flow is complex and irregular in this system, accompanied with rich flow structures and large temperature and velocity fluctuations. Under the application of cusp magnetic field, the turbulent melt flow is significantly inhibited with the simple flow structure as well as the periodic temperature and velocity fluctuations. The anticlockwise periodical flow in the melt induced by cusp magnetic field can help to reduce the impurity in the crystal and suppress the crystal/melt interface deformation. In particular, the thermal waves on the melt free surface are firstly observed with cusp magnetic field for large-size Czochralski silicon crystal growth, traveling in the direction of crucible rotation.

In situ observation with x-ray for tentative exploration of laser beam welding processes for aluminum-based alloys

In recent years, laser processes have taken an ever-increasing market share in the manufacture of components. The development of new, improved beam sources with corresponding systems technology and the decreasing investment costs of the beam sources are important keys to this success. Particularly, high frequency beam oscillation has great potential in laser beam welding and cutting. The main obstacle for the widespread breakthrough of high frequency (HF) beam oscillation is the still insufficient understanding of the underlying physical mechanisms. Gaining a deeper insight is essential for process optimization. The in situ observation with x rays enables the visualization and analysis of these highly dynamic processes inside the workpiece. The goal of the performed experiment described in this paper was to in situ analyze the structural evolution of and defect generation in laser welding beads of different aluminum alloys. A fiber laser (max. 600 W, cw output power) including a beam scanner control system for rapid beam guidance was used. Of general interest was the comparison between static and oscillated beam guidance and the effects on the joining procedure. This paper shows the initial results of the analysis of the melt pool behavior and seam formation as well as the formation of seam irregularities during the laser process. In the simplest case, radiographs were taken, i.e., 2D projections of the x-ray absorption coefficient distribution within a material. Thereby, recordings from 10 000 up to 40 000 fps could be generated. Furthermore, tomoscopies—the continuous acquisition of tomographic (3D) images, up to 100 tomograms per second—could be generated with proven equipment, whose main components are a high-speed rotation stage and a camera system. The findings will help to get a better understanding of keyhole phenomena as well as effects of turbulent melt flow such as pore formation and guide to solutions for preventing them. Hence, initial results of high frequency beam oscillation processes including melt pool degassing and porosity reduction will be shown and discussed.

Numerical Investigation of Effect of Temperature Profile Imposed on the Crucible Surface on Oxygen Incorporated at the Crystal Melt Interface for 450 mm Diameter Silicon Single Crystal Growth in Presence of CUSP Magnetic Field Using Czochralski Technique

Numerical investigation has been carried out to study the effect of thermal boundary condition on turbulent melt flow and oxygen transfer inside a crucible used to grow 450 mm diameter silicon single crystal using Czochralski method. Two types of thermal boundary conditions, namely, isothermal crucible surface and experimentally measured temperatures on the crucible surface have been imposed on the crucible wall and crucible bottom. Melt motion owing to buoyancy, surface tension variation at the free melt surface as well as crystal and crucible rotation has been considered. Effect of CUSP magnetic field has also been investigated. Melt having aspect ratio of 1 and 0.5, representing high and moderate level of melt inside the crucible have been considered. K-epsilon turbulent model with low Reynolds number formulation has been used for turbulence modelling. Melt motion governed by natural convection, Marangoni convection and crystal as well as crucible rotation shows higher oxygen concentration at melt crystal interface for aspect ratio of 1, when experimental temperature profile is imposed at the crucible wall. For melt aspect ratio of 0.5, natural convection and Marangoni convection dependent melt show higher oxygen concentration for isothermal crucible wall boundary condition. Marangoni convection in absence of rotation of crystal and crucible leads to low oxygen concentration zone near the crystal outer surface, the size of which is larger for melt aspect ratio of 0.5. Maximum turbulent viscosity values for aspect ratio of 1, in case of experimental temperature at the crucible are higher compared to isothermal case, which is just the opposite of the trend for aspect ratio of 0.5. Imposing a CUSP magnetic field leads to similar distribution of turbulent viscosity for both types of thermal boundary conditions, irrespective of the type of thermal boundary condition. Oxygen concentration at the melt crystal interface is found to be higher in presence of CUSP magnetic field. Value of maximum turbulent viscosity for the two boundary condition are similar for melt aspect ratio of 1.0. Crucible surface temperature profile for aspect ratio of 1, on actual melt aspect ratio of 0.5 predicts lower oxygen concentration at melt crystal interface. Values and distribution of turbulent viscosity is however the same for both temperature profiles.

Potential of using inert gas flows for controlling the quality of as-grown silicon single crystal

We have improved the well-known Czochralski single crystal silicon growth method by using two argon gas flows. One flow is the main one (15–20 nl/min) and is directed from top to bottom along the growing single crystal. This flow entrains reaction products of melt and quartz crucible (mainly SiO), removes them from the growth chamber through a port in the bottom of the chamber and provides for the growth of dislocation-free single crystals from large weight charge. Similar processes are well known and have been generally used since the 1970s world over. The second additional gas flow (1.5–2 nl/min) is directed at a 45 arc deg angle to the melt surface in the form of jets emitted from circularly arranged nozzles. This second gas flow initiates the formation of a turbulent melt flow region which separates the crystallization front from oxygen-rich convective flows and accelerates carbon evaporation from the melt. It has been confirmed that oxygen evaporated from the melt (in the form of SiO) acts as transport agent for nonvolatile carbon. Commercial process implementation has shown that carbon content in as-grown single crystals can be reduced to below the carbon content in the charge. Single crystals grown with two argon gas flows have also proven to have highly macro- and micro-homogeneous oxygen distributions, with much greater lengths of single crystal portions in which the oxygen concentration is constant and below the preset limit. Carbon contents of 5–10 times lower than carbon content in the charge can be achieved with low argon gas consumption per one growth process (15–20 nl/min vs 50–80 nl/min for conventional processes). The use of an additional argon gas flow with a 10 times lower flowrate than that of the main flow does not distort the pattern of main (axial) flow circumvention around single crystal surface, does not hamper the “dislocation-free growth” of crystals and does not increase the density of microdefects. This suggests that the new method does not change temperature gradients and does not produce thermal shocks that may generate thermal stresses in single crystals.

Filler metal distribution and processing stability in laser-arc hybrid welding of thick HSLA steel

Welds made by high power laser beam have deep and narrow geometry. Addition of filler wire by the arc source, forming the laser-arc hybrid welding (LAHW) process, is very important to obtain required mechanical properties. Distribution of molten wire throughout the entire weld depth is of concern since it tends to have low transportation ability to the root. Accurate identification of filler metal distribution is very challenging. Metal-cored wires can provide high density of non-metallic inclusions (NMIs) which are important for acicular ferrite nucleation. Accurate filler distribution can be recognized based on statistical characterization of NMIs in the weld. In the present study, it was found that the amount of filler metal decreased linearly towards the root. The filler metal tends to accumulate in the upper part of the weld and has a steep decrease at 45–55 % depth which also has wavy pattern based on longitudinal cuts. Substantial hardness variation in longitudinal direction was observed, where in the root values can reach > 300 HV. Excessive porosity was generated at 75 % depth due to unstable and turbulent melt flow based on morphology of prior austenite grains. The delicate balance of process parameters is important factor for both process stability and filler metal distribution.

Opportunity to use inert gas flow for control of qualitative characteristics of the grown silicon single crystals

The process of growing silicon single crystals by the Czochralski method has been improved, which involves the use of two argon streams. 1st, the main flow, 15—20 nl/min, is directed from top to bottom along the growing single crystal. It captures the reaction products of the melt with a quartz crucible (mainly SiO), removes them from the chamber through a nozzle in the lower part of the chamber and provide dislocation-free single crystals from large loads. Similar processes are known and widely used in world practice since the 1970s. 2nd, additional flow, 1.5—2 nl/min, is directed at an angle of 45° to the surface of the melt in the form of jets from nozzles arranged in a ring. This flow initiates the formation of a region of turbulent melt flow, which isolates the crystallization front from convective flows enriched with oxygen, and also enhances the evaporation of carbon from the melt. It is confirmed that the oxygen evaporated from the melt (in the form of SiO) is a «transport» for non-volatile carbon. Carrying out industrial processes showed that the carbon content in the grown single crystals can be significantly reduced, up to values smaller than in the feedstock. In single crystals grown using two argon streams, an increased macro- and micro-uniformity of the oxygen distribution, a significantly larger crystal length with a given, constant oxygen concentration, were also recorded. Achieving a carbon concentration of 5 to 10 times less than in the feedstock is possible with small amounts of argon for melting (15—20 nl/min compared to 50—80 nl/min used in conventional processes. The use of an additional argon flow, which has an outflow intensity 10 times lower than that of the main flow, does not distort the nature of the flow around the single crystal surface (“axial”), does not disrupt the growth of a dislocation-free single crystal, does not increase the density of microdefects, which indicates the absence of changes in temperature gradients and thermal shock leading to thermal stresses in a single crystal.

Transient global modeling for the pulling process of Czochralski silicon crystal growth. I. Principles, formulation, and implementation of the model

Abstract A transient global model for the crystal pulling process was developed for magnetic-field-applied Czochralski silicon (CZ-Si) growth. Heat transfer by solid conduction, melt convection, and diffuse gray radiation is taken into account for the crystal, melt, and other components in the furnace. The mesh adaption and view factor updating for the dynamic pulling process were realized by the structured grid deformations for different domains. By imposing the thermal boundary conditions in the vicinity of the triple point, the inverse control by the virtual proportional integral derivative controller and heat flux is acceptable for the transient global simulation of the pulling process in CZ-Si crystal growth. The applied cusp-shaped magnetic field (CMF) suppressed the turbulent melt flow and stabilized the heat and mass transport. CMF with different zero-Gauss-plane locations resulted in different flow patterns, which could affect the impurity transport during the pulling process. This developed transient global model can be applied for the segregation predictions of impurities (oxygen and carbon) and dopants in the CZ-Si growing process.

Optimal Control of Semiconductor Melts by Traveling Magnetic Fields

In this paper, the optimal control of traveling magnetic fields in a process of crystal growth from the melt of semiconductor materials is considered. As controls, the phase shifts of the voltage in the coils of a heater-magnet module are employed to generate Lorentz forces for stirring the crystal melt in an optimal way. By the use of a new industrial heater-magnet module, the Lorentz forces have a stronger impact on the melt than in earlier technologies. It is known from experiments that during the growth process, temperature oscillations with respect to time occur in the neighborhood of the solid-liquid interface. These oscillations may strongly influence the quality of the growing single crystal. As it seems to be impossible to suppress them completely, the main goal of optimization has to be less ambitious, namely one tries to achieve oscillations that have a small amplitude and a frequency which is sufficiently high such that the solid-liquid interface does not have enough time to react to the oscillations. In our approach, we control the oscillations at a finite number of selected points in the neighborhood of the solidification front. The system dynamics is modeled by a coupled system of partial differential equations that account for instationary heat conduction, turbulent melt flow, and magnetic field. We report on numerical methods for solving this system and for the optimization of the whole process. Different objective functionals are tested to reach the goal of optimization.

Collision of Micro-sized Non-metallic Inclusions in Liquid Steel Flows: A Computational Study

The quality of steel is limited, among others, by the contained non-metallic inclusions. A key factor in this context is the growth of micro-sized indigenous inclusions. Those are assumed to form larger inclusions or clusters by collision in the turbulent melt flow. In this study, a numerical model is established to investigate the collision probability of spherical and clustered non-metallic inclusions in steel by discrete particle simulations. A strong dependence of the agglomeration probability from the particle sizes, form and shear rate in the process is observed. Larger particles are hampered to coagulate because of the significant influence of the lubrication force. Thus, the maximum particle size of spherical particles is limited, while clustered agglomerates can still grow. From the simulation results, models for the determination of the collision coefficient are deduced for spherical particles and non-spherical particle clusters.

Do rotating magnetic fields unconditionally lead to grain refinement? A case study for directionally solidified Al-10wt%Cu alloys

The effect of solidification velocity and electromagnetic stirring on grain refining was investigated experimentally during the directional solidification of rod-like Al-10wt%Cu alloy samples. Applying low solidification velocities leads to a dendritic microstructure consisting of elongated equiaxed crystals, which result from fragmented dendrite arms forming new grains. This grain-refining effect vanishes for higher solidification velocities, leading to a microstructure dominated by a lower number of larger columnar grains. Moderate electromagnetic stirring under laminar flow conditions does not promote grain refinement. By contrast, a sufficiently strong forced melt flow induced by a rotating magnetic field significantly increases the number of grains in the range of solidification velocities investigated within this study. It is assumed that a turbulent melt flow supports the fragmentation of dendrite arms and thus the formation of new grains, which finally leads to grain refinement.

Influence of melt flow in the gating system on microstructure and mechanical properties of high pressure die casting AZ91D magnesium alloy

The influence of melt flow in the gating system on microstructure and mechanical properties of high pressure die casting AZ91D magnesium alloy was investigated. Experimental and theoretical results showed that the externally solidified crystals (ESCs) were broken and eliminated by the shear stress caused by the turbulent melt flow in gating system. In comparison with the casting fabricated with conventional gating system, a simultaneous increase in both strength (~ 16% higher ultimate tensile strength) and ductility (~ 70% higher elongation) were achieved in the casting fabricated with modified gating system. Furthermore, different modes of crack initiation and propagation were found. For the one fabricated with the conventional gating system, the crack initiated at the shrinkage pore and propagated in a quasi-cleavage fracture form. In contrast, for the other one fabricated with the modified gating system, micro-voids appeared between the α-Mg grains and β-Mg17Al12 phase, and crack propagated in a ductile form.