SHEAR INDUCED GRAIN REFINEMENT OF A CAST INGOT

TECHNICAL FIELD

The present disclosure relates to metal casting generally and more specifically to controlling delivery of molten metal to a mold cavity.

SUMMARY

Wide variation of grain size can be found through the cross-section of a Direct-Chill (DC) cast ingot due to the location-dependent solidification rate inherent to the process. The use of a turbulent jet as a metal entrance method has the potential to significantly reduce grain size and its variability upon location. Experiments have been conducted investigating the influence of jet power on the grain size and distribution in Al4.5Cu DC cast ingots. The findings indicate that significantly increasing the jet power may not appreciably decrease the grain size. Instead, a threshold jet power required for grain refinement has been determined, beyond which only marginal improvements are anticipated.

BRIEF DESCRIPTION OF THE DRAWINGS

The specification makes reference to the following appended figures, in which use of like reference numerals in different figures is intended to illustrate like or analogous components. The elements included in the illustrations herein may not be drawn to scale.

FIG. 1 is a top view schematic diagram depicting the location of samples taken for metallographic analysis from an ingot according to certain aspects of the present disclosure.

FIG. 2 is a surface plot depicting grain size distribution in one quadrant of a standard-cast (SD cast) Al4.5Cu alloy.

FIG. 3 is a surface plot depicting grain size distribution in one quadrant of a jet-cast (JT cast) Al4.5Cu alloy having a jet with a Reynolds number of 64,000.

FIG. 4 is a surface plot depicting grain size distribution in one quadrant of a jet-cast Al4.5Cu alloy having a jet with a Reynolds number of 69,000.

FIG. 5 is a surface plot depicting grain size distribution in one quadrant of a jet-cast Al4.5Cu alloy having a jet with a Reynolds number of 81,000.

FIG. 6 is a surface plot depicting grain size distribution in one quadrant of a jet-cast Al4.5Cu alloy having a jet with a Reynolds number of 97,000.

FIG. 7 is a surface plot depicting grain size distribution in one quadrant of a jet-cast Al4.5Cu alloy having a jet with a Reynolds number of 121,000.

FIG. 8 is a surface plot depicting spatial secondary dendrite arm spacing in one quadrant of a standard-cast (SD cast) Al4.5Cu alloy.

FIG. 9 is a surface plot depicting spatial secondary dendrite arm spacing in one quadrant of a jet-cast (JT cast) Al4.5Cu alloy having a jet with a Reynolds number of 64,000.

FIG. 10 is a surface plot depicting spatial secondary dendrite arm spacing in one quadrant of a jet-cast Al4.5Cu alloy having a jet with a Reynolds number of 69,000.

FIG. 11 is a surface plot depicting spatial secondary dendrite arm spacing in one quadrant of a jet-cast Al4.5Cu alloy having a jet with a Reynolds number of 81,000.

FIG. 12 is a surface plot depicting spatial secondary dendrite arm spacing in one quadrant of a jet-cast Al4.5Cu alloy having a jet with a Reynolds number of 97,000.

FIG. 13 is a surface plot depicting spatial secondary dendrite arm spacing in one quadrant of a jet-cast Al4.5Cu alloy having a jet with a Reynolds number of 121,000.

FIG. 14 is a chart depicting average grain size and dendrite arm spacing as a function of the jet Reynolds number (Re_j).

FIG. 15 is a chart depicting the spread (e.g., range) of grain sizes and dendrite arm spacing as a function of the jet Reynolds number (Re_j).

FIG. 16 is a series of micrographs depicting samples taken from ingots cast using a jet-cast technique and a standard-cast technique.

FIG. 17 is a partial cross-sectional view of a metal casting system with a single nozzle according to certain aspects of the present disclosure.

FIG. 18 is a partial cross-sectional view of a metal casting system with multiple nozzles according to certain aspects of the present disclosure.

DETAILED DESCRIPTION

Certain aspects and features of the present disclosure relate to the application of a turbulent mixing jet as a method to homogenize and refine the grain structures found within Direct-Chill (DC) cast aluminum ingots. By leveraging an understanding of convective solidification, the influence of jet (e.g., shear) power on grain refinement and homogeneity in cast aluminum products, such as Al4.5Cu rolling slab ingots, can be examined. Through experimentation described herein, it has been found that surprisingly desirable metallurgical characteristics can be achieved in a Direct-Chill (DC) cast product by supplying the liquid metal to the mold and liquid sump using a nozzle designed to provide high-velocity flow. Such a nozzle can have a reduced diameter opening, thereby providing higher velocity flow at a constant volumetric flow rate as compared to a nozzle with a standard diameter opening.

A fine and uniform grain structure is desirable for optimum formability and homogenous mechanical properties in wrought aluminum products. Grain structure (size, distribution, and morphology) is an important parameter influencing defects, such as hot cracking. In Direct-Chill (DC) castings, the grain structure depends on numerous factors, including the alloy composition, the introduction of heterogeneous nucleation sites (e.g., grain refiner), growth conditions, and cooling rate. Due to the shape of the solidifying interface (e.g., sump) formed during DC casting, the solidification rate of an individual grain is extremely position dependent. This variation in solidification rates can lead to large variances in grain size and structure through large rolling slab ingots.

In general, grain morphology in commercial, grain-refined castings is equiaxed and dendritic. However, non-dendritic grains have been reported in DC cast aluminum alloys upon grain-refining. It has been reported that within the central portion of a grain refined AA2024 alloy, isothermal dendrites (“floating grains”) appear to be non-dendritic compared to the rest of the structure, which is dendritic.

In an attempt to modify macrosegregation patterns, a turbulent jet outfitted with a special mixing nozzle was used to mobilize grains which had sedimented to the bottom of the sump. The turbulent stirring caused non-dendritic grains to be found, and an increase in grain refinement and homogeneity of the solidified structure was observed. Similar to these results, in electromagnetic casting (EMC), unlike conventional DC casting, the grain size is more homogeneously distributed throughout the cross section of the ingot, which is a result of forced convection, lower thermal gradients, and more intensive transport of the solid phase within the slurry zone (e.g., “mushy zone”).

The formation of non-dendritic structures is beneficial for the structural homogeneity of DC cast products, improving their mechanical properties in the vicinity of the solidus, diminishing macrosegregation, and decreasing cracking susceptibility. On the other hand, it had been thought that the formation of very fine globular grains can increase the hot tearing susceptibility due to the limited permeability of the semi-solid region (i.e., the “mushy zone”). An important feature of the non-dendritic grains that are formed during solidification is the unique dependence of the non-dendritic grain size on the cooling rate. The size of a non-dendritic grain at a certain cooling rate is the same as the dendrite arm spacing of a dendritic grain formed at the same cooling rate.

These observations may be non-unique to DC casting. The viscosity of partially solidified melts had been investigated using a couette viscometer. Unintentionally, the dendritic structure had been sheared and broken up and it was found that the partially solidified material exhibited thixotropic behavior. This discovery led to rheo-casting, an extrusion and die casting technique which leverages the properties and unique microstructure of the sheared and broken up dendritic structure. This discovery has spurred a research interest in solidification under forced convection. In an attempt to put the present disclosure into perspective, a highlight of this research is presented.

Solidification Behavior Under Forced Convection.

Nearly all alloys of commercial importance solidify dendritically, with either a columnar or an equiaxed dendritic structure. During dendritic solidification of castings and ingots, a number of processes take place simultaneously within the semi-solid region (e.g., “mushy zone”). These processes include crystallization, solute redistribution, ripening, interdendritic fluid flow, and solid movement. The dendritic structure is greatly affected by the interdendritic flow and solid movement, which, in conventional solidification, is caused by internal factors such as density difference and heterogeneous distribution of temperature.

In studying conventional solidification, transparent organic alloys have been intensively used to investigate the solidification behavior at microstructural level by direct observation. However, this may not be possible for solidification under forced convection, even using organic analogue, due to the blurred image caused by intensive stirring. For this reason, current understanding of the solidification behavior under forced convention is obtained indirectly by examination of the final solidified microstructures.

It has been established from experimental observations that solidification under melt stirring produces non-dendritic structures. Work on the Sn—Pb system using rotational rheometers confirmed that the solid phase in the semisolid state has either a degenerated dendritic structure or rosette morphology. With prolonged stirring time, such particles change to a more or less spherical morphology containing entrapped liquid by a ripening process. Increasing the shear rate accelerates this morphological transition and reduces the amount of entrapped liquid inside the solid particles. The rosette morphology of solid particles has also been observed in many stirred alloys by other investigators using rod and impeller types of stirrer. Later work on solidification under magnetohydrodynamic (MHD) stirring confirmed the formation of a fine and degenerated dendritic structure.

Grain Density.

It has been observed that under forced convection, larger grain densities are present than under conventional solidification techniques. The grain refinement is attributed to the fact that the convection associated with the mold filling remains strong as solidification commences. Experiments had been performed which showed that if convection is present at the onset of solidification, a structure with many smaller grains may form. Superheated liquid Al4.5Cu alloy was drawn into a thin-plate copper mold. The dendritic structure on the surface of the cast plates were visible to the naked eye. If metal was drawn into the mold with a large amount of superheat, large dendrites were observed on the surface of the cast plates. If the metal was drawn with significantly less superheat, a far greater number of smaller dendrites were observed. It was believed that during an experiment, the tip of the metal in the channel would freeze first, stopping flow. For metal cast with a higher superheat, the metal behind the frozen flow tip would be quiescent when it began to solidify. Since convection was not present during solidification, the final grain size was large. For the case of metal cast with significantly less superheat, it was likely that the metal behind the flow tip already began to solidify before the flow tip froze, which resulted in a finer grain size throughout the casting.

One theory to explain the effect of convection on grain refinement was tested by using carbon tetrabromide with salol, a transparent analog casting system, and video equipment to show that when convection was present during solidification, many solid particles would suddenly be emitted from the “mushy zone” and would move into the bulk melt. The “big bang” theory explained this observation by postulating that convection caused thermal fluctuations which in turn lead to fluctuations in growth rate and re-melting of dendrite arms. The detached arms were then thought to be transported by convective flows or buoyancy into the bulk melt, resulting in a structure with many smaller grains.

To explain the observed grain refinement by melt stirring, a dendrite arm fragmentation mechanism was proposed to account for grain multiplication. It was suggested that dendrite arms bend plastically under the shear force created by melt stirring. Plastic bending introduces large misorientations into the dendrite arms in the form of “geometrically necessary dislocations.” At high temperature, such dislocations rearrange themselves to form high angle grain boundaries through recrystallisation. Any grain boundary with an energy greater than twice the solid/liquid interfacial energy is then wetted by liquid metal, resulting in the detachment of dendrite arms.

Following these early suggestions, it was suggested that secondary dendrite arms can detach at their roots because of re-melting due to solute enrichment and thermosolutal convection. To explain the crystal multiplication in semisolid processing, it was suggested that temperature fluctuations in the MHD rheocasting process play a significant role in the structural evolution. A continuous nucleation might take place in the absence of a distinct recalescence where each volume element of the liquid passes periodically through different temperature zones.

The dendrite fragmentation mechanism attempts to rationalize the final microstructural features observed in the solid, but the important remaining question is how likely is it that shearing can exert such a high bending moment to small dendrite arms to fracture them. Under some theories, the microscale of turbulence has to be of the order of particle size for the viscous forces to be active on bending the dendrite arms, and that is possible only at a very high shear rate. Moreover, fragmented dendrite arms are expected to grow dendritically in the melt, at least during the initial period of growth, until impingement of diffusion fields occur. This understanding is not compatible with some experimental observations, where primary particles can be seen to be few in number but exhibit a degenerated dendritic or spherical microstructure.

In light of some of the difficulties with existing dendrite fragmentation theory, the experimentally observed grain refinement under high degrees of convective turbulence may be explained by a copious nucleation mechanism. Under intensive mixing action, both temperature and composition fields inside the liquid alloy are extremely uniform. During continuous cooling under forced convection, heterogeneous nucleation takes place at the same time throughout the whole liquid phase. Compared with conventional solidification, the actual nucleation rate may not be increased, but all the nuclei formed will survive due to the uniform temperature field, resulting in an increased effective nucleation rate. In addition, the intensive mixing action will disperse the clusters of potential nucleation agents, giving rise to an increased number of potential nucleation sites.

Experimental Apparatus and Procedure

For each experiment, charges were loaded into a commercial gas burner furnace. The melt was degassed and filtered to commercial standards. The charge was inoculated with approximately 25 pμm TiB grain refiner in order to maintain consistency with previous investigations. A typical “coffin type” bottom block was used along with a 600×1750 mm WAGSTAFF LHC™ open top mold. The steady state casting speed was approximately 65 mm/min. The casting speed and metal level were varied during approximately the first 500 mm of the cast before reaching steady state values. The alloy used was an Al4.5Cu alloy, although similar or corresponding results can be achieved with other alloys.

To vary the degree of jet power utilized to refine grain structure, a series of fused silica downspouts were prepared, each with a unique diameter. The unique diameter, along with the flow rate and physical properties of the Al4.5Cu alloy, led to a unique turbulent Reynolds number for each trial. The Reynolds numbers investigated during this series of experiments were: 64,000, 69000, 81,000, 97,000, and 121,000. By comparison, the Reynolds number associated with a standard nozzle may be at or less than 15,000. To maintain the intensity of the mixing jet, no combination bag was used during these trials. In some cases, the use of a combination bag, which is often standard, may reduce the efficacy of any jet created by the nozzle.

Following casting, the cold ingots were each cross sectioned at approximately 1800 mm of cast length. Each cross section was divided into quadrants by symmetry, and a series of 45 samples were removed using a 1-inch core drill for metallographic analysis, as depicted in FIG. 1.

FIG. 1 is a top view schematic diagram depicting the location of samples 102 taken for metallographic analysis from an ingot 100 according to certain aspects of the present disclosure. Notations A, C, and E correspond to micrograph images depicted in FIG. 16. The ingot 100 can be considered as having four quadrants, including quadrants 1 (not labeled), quadrant 2, quadrant 3, and quadrant 4.

The metallographic samples were etched with a dilute solution of hydrofluoric acid and analyzed for grain size and dendrite arm spacing using an optical microscope according to the established line-intercept method.

FIGS. 2-7 are surface plots depicting grain size distribution in one quadrant of cast ingots prepared using standard cast techniques or various jet stirring techniques (e.g., high-velocity jet techniques or jet-cast ingots) according to certain aspects of the present disclosure. The surface plots can be taken at a length of approximately 1800 mm along the cast length of the ingots. The surface plots depict grain size distributions using color bars at an identical scale, ranging from approximately 50 micrometers (e.g., dark blue) up to at or over 250 micrometers (e.g., dark red). Generally, the grain size can vary from about 50 μm at the short face (e.g., far left of the plot), where solidification is most rapid, to over 250 μm near the center where the sump is steepest and solidification rates are the slowest.

FIG. 2 is a surface plot depicting grain size distribution in one quadrant of a standard-cast (SD cast) Al4.5Cu alloy. The standard cast alloy can be performed without a high-velocity jet (e.g., with a jet having a Reynolds number at or below 15,000, or with a combination bag). The range of grain sizes seen in the standard-cast ingot ranged from approximately 50 micrometers to at or above approximately 250 micrometers. The region of approximately −500 to 0 mm from center along the x axis and approximately 50 to 150 mm from center along the y axis shows large grain size distributions (e.g., distributions of large grains). The region near the bottom left corner of the surface plot shows small grain size distributions (e.g., distribution of smaller grains). For reference, the region at the bottom left corner of the surface plot represents the region near the middle of the short face of the ingot.

FIG. 3 is a surface plot depicting grain size distribution in one quadrant of a jet-cast (JT cast) Al4.5Cu alloy having a jet with a Reynolds number of 64,000. The regions near the edges of the plot show smaller grain size distributions (e.g., distributions of smaller grains) than regions near the middle of the plot.

FIG. 4 is a surface plot depicting grain size distribution in one quadrant of a jet-cast Al4.5Cu alloy having a jet with a Reynolds number of 69,000. The regions near the edges of the plot show smaller grain size distributions (e.g., distributions of smaller grains) than regions near the middle-right of the plot.

FIG. 5 is a surface plot depicting grain size distribution in one quadrant of a jet-cast Al4.5Cu alloy having a jet with a Reynolds number of 81,000. The regions near the edges of the plot show smaller grain size distributions (e.g., distributions of smaller grains) than regions near the middle-right of the plot.

FIG. 6 is a surface plot depicting grain size distribution in one quadrant of a jet-cast Al4.5Cu alloy having a jet with a Reynolds number of 97,000. The regions near the edges of the plot show smaller grain size distributions (e.g., distributions of smaller grains) than regions near the middle-right of the plot.

FIG. 7 is a surface plot depicting grain size distribution in one quadrant of a jet-cast Al4.5Cu alloy having a jet with a Reynolds number of 121,000. The regions near the edges of the plot show smaller grain size distributions (e.g., distributions of smaller grains) than regions near the middle-right of the plot.

With respect to FIGS. 2-7, it is apparent that the grain size pattern from each of the jet-cast ingot trials appears to be fairly similar in its distortion of the standard result. Regardless of the amount of jet power (e.g., Reynolds number), a trend towards slower cooling—and thus larger grain sizes—towards the center of the ingot is observed similar to that observed in the standard-cast trials. However, the range of grain sizes observed is greatly reduced. While the standard-cast ingot had grain sizes up to at or above approximately 250 micrometers, the grain sizes of the jet-cast ingots were much smaller, with the majority of grains at approximately 100 micrometers and the largest grains at, around, or under 150 micrometers in diameter.

FIGS. 8-13 are surface plots depicting the spatial secondary dendrite arm spacing (DAS) profile in one quadrant of cast ingots prepared using standard cast techniques or various jet stirring techniques (e.g., high-velocity jet techniques or jet-cast ingots) according to certain aspects of the present disclosure. The surface plots can be taken at a length of approximately 1800 mm along the cast length of the ingots. The surface plots depict dendrite arm spacing using color bars at an identical scale, ranging from approximately 20 micrometers (e.g., dark blue) up to at or over 70 micrometers (e.g., dark red). The ingots used for the surface plots of FIGS. 8-13 can be the same ingots used for respective surface plots of FIGS. 2-7.

FIG. 8 is a surface plot depicting spatial secondary dendrite arm spacing in one quadrant of a standard-cast (SD cast) Al4.5Cu alloy. The standard cast alloy can be performed without a high-velocity jet (e.g., with a jet having a Reynolds number at or below 15,000, or with a combination bag). The range of grain sizes seen in the standard-cast ingot ranged from approximately 50 micrometers to at or above approximately 250 micrometers. The region of approximately −500 to 0 mm from center along the x axis and approximately 50 to 150 mm from center along the y axis shows large grain size distributions (e.g., distributions of large grains). The region near the bottom left corner of the surface plot shows small grain size distributions (e.g., distributions of smaller grains). For reference, the region at the bottom left corner of the surface plot represents the region near the middle of the short face of the ingot.

FIG. 9 is a surface plot depicting spatial secondary dendrite arm spacing in one quadrant of a jet-cast (JT cast) Al4.5Cu alloy having a jet with a Reynolds number of 64,000. The regions near the edges of the plot show smaller grain size distributions (e.g., distributions of smaller grains) than regions near the middle of the plot.

FIG. 10 is a surface plot depicting spatial secondary dendrite arm spacing in one quadrant of a jet-cast Al4.5Cu alloy having a jet with a Reynolds number of 69,000. The regions near the edges of the plot show smaller grain size distributions (e.g., distributions of smaller grains) than regions near the middle-right of the plot.

FIG. 11 is a surface plot depicting spatial secondary dendrite arm spacing in one quadrant of a jet-cast Al4.5Cu alloy having a jet with a Reynolds number of 81,000. The regions near the edges of the plot show smaller grain size distributions (e.g., distributions of smaller grains) than regions near the middle-right of the plot.

FIG. 12 is a surface plot depicting spatial secondary dendrite arm spacing in one quadrant of a jet-cast Al4.5Cu alloy having a jet with a Reynolds number of 97,000. The regions near the edges of the plot show smaller grain size distributions (e.g., distributions of smaller grains) than regions near the middle-right of the plot.

FIG. 13 is a surface plot depicting spatial secondary dendrite arm spacing in one quadrant of a jet-cast Al4.5Cu alloy having a jet with a Reynolds number of 121,000. The regions near the edges of the plot show smaller grain size distributions (e.g., distributions of smaller grains) than regions near the middle-right of the plot.

With respect to FIGS. 8-13, it is apparent that the maximum spatial secondary dendrite arm spacing from each of the jet-cast ingot trials appears to be substantially lower than that of the standard-cast ingot. Further, for the jets of the lowest Reynolds numbers (e.g., 64,000; 69,000; and 81,000 of FIGS. 9-11, respectively), the range of DAS is much smaller, with the periphery being approximately 35-40 micrometers and growing to approximately 50 micrometers in the center. The DAS is also much more homogenous for these casts than the standard or higher Reynolds number jets. For these jets of higher Reynolds number (e.g., 97,000 and 121,000 of FIGS. 12 and 13, respectively), the periphery of the ingot exemplifies DAS of only approximately 25 micrometers and grows to perhaps 50 micrometers in the center. While the profiles may be similar to the standard case in some respects, the average DAS is smaller for these casts than the other samples.

FIG. 14 is a chart depicting average grain size and dendrite arm spacing as a function of the jet Reynolds number (Re_j). In order to better quantify the effects of turbulent jet stirring, the standard distribution bag dimensions and velocities specified to generate an effective Reynolds number of 15,000 for the standard case were utilized.

There appears to be little influence of the power of the jet on the respective dimension. The simple addition of the mixing jet realized an approximate 25% reduction in grain size for all mixing jets. The DAS was slightly less responsive, only exhibiting an approximate 10% reduction for the most powerful jets (e.g., 97,000 and 121,000), while the other mixing jets exhibited smaller deviation.

Certain aspects and features of the present disclosure can result in a cast product having an average grain size at or below approximately 150 μm, 149 μm, 148 μm, 147 μm, 146 μm, 145 μm, 144 μm, 143 μm, 142 μm, 141 μm, 140 μm, 139 μm, 138 μm, 137 μm, 136 μm, 135 μm, 134 μm, 133 μm, 132 μm, 131 μm, 130 μm, 129 μm, 128 μm, 127 μm, 126 μm, 125 μm, 124 μm, 123 μm, 122 μm, 121 μm, 120 μm, 119 μm, 118 μm, 117 μm, 116 μm, 115 μm, 114 μm, 113 μm, 112 μm, 111 μm, 110 μm, 109 μm, 108 μm, 107 μm, 106 μm, 105 μm, 104 μm, 103 μm, 102 μm, 101 μm, or 100 μm.

FIG. 15 is a chart depicting the spread (e.g., range) of grain sizes and dendrite arm spacing as a function of the jet Reynolds number (Re_j). Again, the standard-cast ingot is identified with a Reynolds number of 15,000. The spread of DAS is not affected much by the introduction of a turbulent jet. However, the grain size exhibited a drastic change. The standard-cast ingot had a grain size distribution of approximately 200 micrometers (e.g., from approximately 100 micrometers to approximately 300 micrometers), while the introduction of a turbulent jet reduced this range to only approximately 75 micrometers (e.g., from approximately 100 micrometers to approximately 175 micrometers). There seemed to be little impact of jet power on this range, with the simple presence of the jet immediately reducing the grain size distribution.

Certain aspects and features of the present disclosure can result in a cast product having a maximum grain size at or below approximately 290 μm, 285 μm, 280 μm, 275 μm, 270 μm, 265 μm, 260 μm, 255 μm, 250 μm, 245 μm, 240 μm, 235 μm, 230 μm, 225 μm, 220 μm, 215 μm, 210 μm, 205 μm, 200 μm, 195 μm, 190 μm, 185 μm, 180 μm, 175 μm, 170 μm, 165 μm, 160 μm, 155 μm, 150 μm, 145 μm, 140 μm, 135 μm, or 130 μm.

Certain aspects and features of the present disclosure can result in a cast product having a grain size spread (e.g., range between minimum and maximum grain sizes) at or less than approximately 200 μm, 195 μm, 190 μm, 185 μm, 180 μm, 175 μm, 170 μm, 165 μm, 160 μm, 155 μm, 150 μm, 145 μm, 140 μm, 135 μm, 130 μm, 125 μm, 120 μm, 115 μm, 110 μm, 105 μm, 100 μm, 95 μm, 90 μm, 85 μm, 80 μm, 75 μm, 70 μm, 65 μm, 60 μm, or 55 μm. In other words, a cast product having a smaller grain size spread can be considered to have a more homogenous grain size distribution, whereas a cast product having a larger grain size spread can be considered to have a less homogenous grain size distribution.

FIG. 16 is a series of micrographs depicting samples taken from ingots cast using a jet-cast technique and a standard-cast technique. The jet-cast technique was performed using a jet having a Reynolds number of approximately 97,000. The samples were taken at locations A, C, and E of the ingot of FIG. 1. Location A corresponds to a center region of the ingot, C corresponds to a mid-thickness region, and E corresponds to an edge region.

While quantifiable data has been presented herein, there are other qualitative observations seen in FIG. 16. For the standard case, locations E and C exhibit dendritic structure, with E being much finer than that of C. Sample A for the standard case is seemingly non-dendritic in structure. By contrast, samples A and C for the jet-cast case are characterized by a fine non-dendritic microstructure, while sample E seems to be a mix of non-dendritic and dendritic microstructure, also very fine.

After experiment and trial, it is apparent that there seems to be little impact of the power of a turbulent jet to increase the degree of grain refinement. This is a similar to other observations related to how increasing stirring speed did not influence particle density or size significantly. Instead, the simple presence of stirring promoted grain refinement and non-dendritic structures. Of note, increasing the shear rate may cause the particle density to increase while decreasing the average particle size. It seems that the mixing by a turbulent jet is enough to homogenize temperature isotherms to promote a copious nucleation mechanism. The increased survival rate of the nuclei restricts the growth of the grains and generates a smaller overall grain size.

Of note, the most turbulent jets (e.g., 97,000 and 121,000) generated the most homogenous grain profile, with the smallest homogenous dendrite arm spacing profile. While the lower Reynolds number jets de-stratified the molten pool enough to generate uniform solidification rates (e.g., as apparent by similar microstructure), the most turbulent jets segregated microstructure while generating homogenous grain size. Further, the most turbulent jets from the investigations into macrosegregation were designed to suspend grains from the center and inhibit preferential sedimentation by redistributing “superfluous” floating grains. While this trend seems to be observed by the homogenous grain size, there does seem to be a disconnect from the DAS. One possible explanation is that there is cross-talk between the grain size data and the DAS data because DAS is difficult to isolate in non-dendritic structures. This could artificially skew the distribution of DAS in these jets because they would be the most likely to generate a spheroidal structure (i.e. DAS=Grain Size).

FIG. 17 is a partial cross-sectional view of a metal casting system 1700 with a single nozzle 1708 according to certain aspects of the present disclosure. The metal casting system 1700 can be used to cast a metal product as described herein, such as casting using a nozzle shaped to generate a jet 1734 of molten metal 1726 having a sufficiently high Reynolds numbers, such as those described with reference to FIGS. 3-7 and 9-13. However, in some cases, other casting systems can be used.

A metal source 1702, such as a tundish, can supply molten metal 1726 down a feed tube 1736. A bottom block 1722 may be lifted by a hydraulic cylinder 1724 to meet the walls of the mold cavity 1716. As molten metal begins to solidify within the mold, the bottom block 1722 can be steadily lowered. The cast metal 1712 can include sides 1720 that have solidified, while molten metal 1726 added to the cast can be used to continuously lengthen the cast metal 1712. In some cases, the walls of the mold cavity 1716 define a hollow space and may contain a coolant 1718, such as water. The coolant 1718 can exit as jets from the hollow space and flow down the sides 1720 of the cast metal 1712 to help solidify the cast metal 1712. The ingot being cast can include solidified metal 1730, transitional metal 1728, and molten metal 1726.

Molten metal 1726 can exit the feed tube 1736 at a nozzle 1708 that is submerged in the molten metal 1726. The nozzle 1708 can be a part of the feed tube 1736 or can be a separable part. The nozzle 1708 can have parameters designed to provide flow of the molten metal 1726 (e.g., a jet 1734 of molten metal 1726) at or approximately at a desired Reynolds number.

In some cases, the Reynolds number can be approximated using the equation

$Re = \frac{L_{mold}}{D_{Jet}} C,$

where Re is the Reynolds number, L_moldis the length of the mold at the rolling face of the resultant ingot (e.g., the “Rolling Face” of the ingot 100 depicted in FIG. 1), D_jetis the diameter of the jet, and C is a constant. The constant C can be alloy dependent and can be experimentally determined. The constant C can be a function of mold width (e.g., the length of the mold at the short face of the resulting ingot, such as the “Short Face” of the ingot 100 depicted in FIG. 1), casting speed, kinematic viscosity, and other numerical constants which account for the cross sectional shape of the jet. This example for approximating the Reynolds number is given for casting a rectangular ingot in a rectangular mold. However, one of ordinary skill in the art can calculate a Reynolds number of a jet for casting using molds of other shapes, such as circular billets or non-standard shapes. In some cases, Reynolds number can be a function of jet diameter and effective hydraulic perimeter, or can be otherwise approximated.

The diameter of the jet can be the at or approximately the diameter of the opening of the nozzle (e.g., nozzle 1708). Thus, the Reynold's number for a particular metal casting system (e.g., metal casting system 1700) will decrease as the diameter of the opening of the nozzle (e.g., nozzle 1708) increases and/or as the length of the mold decreases (e.g., decreases towards a round billet the same size as the diameter of the jet). The constants which define the Reynolds number for a particular casting system and alloy can be determined by one of ordinary skill in the art.

Molten metal 1726 exiting the nozzle 1708 can create a jet 1734 of molten metal 1726 having a particular Reynolds number. The desirable characteristics of the jet 1734 as described herein (e.g., to improve metallurgical properties, such as grain refinement) can be achieved with a nozzle 1708 designed or shaped to produce a jet having a desirable Reynolds number (e.g., at or above a threshold number).

FIG. 18 is a partial cross-sectional view of a metal casting system 1800 with multiple nozzles 1807, 1808, 1809 according to certain aspects of the present disclosure. The metal casting system 1800 can be similar to metal casting system 1700 except for the presence of multiple nozzles 1807, 1808, 1809 in place of a single nozzle. The metal casting system 1800 is depicted with three nozzles, although any number of nozzles can be used. In some cases, multiple nozzles 1807, 1808, 1809 can each be supplied from a metal source 1802 by respective, individual feed tubes 1835, 1836, 1837. However, in some cases, multiple nozzles can be fed from a single feed tube (e.g., a branching feed tube). Flow of molten metal 1826 through the multiple nozzles 1807, 1808, 1809 can generate respective jets 1833, 1834, 1835.

The desirable metallurgical effects described herein when metal is cast using a nozzle designed to produce a jet having a Reynolds number at or above a particular threshold Reynolds number (e.g., as described with reference to FIG. 17) can be similarly achieved by using multiple nozzles, wherein the sum of the Reynolds numbers of the multiple jets produced by the multiple nozzles is at or above the particular threshold Reynolds number. In such cases, each of the multiple nozzles may generate a jet having a Reynolds number below the threshold Reynolds number, yet if the sum of the Reynolds numbers of the multiple jets from the multiple nozzles is above that threshold, the desirable metallurgical effects can be obtained. In particular, the sum of the Reynolds numbers of the jets being generated by the multiple nozzles at a particular point in time can be above the threshold Reynolds number. In some cases, a casting system with multiple nozzles may have fewer than all nozzles running simultaneously, such as during different stages of casting (e.g., a start-up phase, a steady state phase, and ending phase). Therefore, as long as the sum of the Reynolds numbers of those jets being generated is at or above the threshold Reynolds number, the desirable results may be achieved.

CONCLUSION

The influence of a turbulent jet on the microstructural properties and distribution of Al4.5Cu rolling slab ingots has been examined. It has been found that the introduction of a turbulent jet promotes grain refinement, while affecting the DAS very little. Even the most turbulent jets do not display a noticeable increase in grain refinement from the least turbulent jet. This seems to indicate that there is a threshold Reynolds number of a turbulent jet which de-stabilizes the isotherms and promotes grain refinement below the lowest evaluated during this investigation. While the influence of the turbulent jet is clear, it is perhaps best suited for a numerical investigation since the effects are so robust and independent of fine fluid-dynamic parameters. It is contemplated that a system for increasing shear in a DC casting system should increase the grain refinement when a simple increase in velocity could not.

Based on the experiments performed herein, it is expected that promoted grain refinement can be achieved in direct chill casting when using a jet having a Reynolds number at or above a threshold Reynolds number of at or approximately 14000, 15000, 16000, 17000, 18000, 19000, 20000, 21000, 22000, 23000, 24000, 25000, 26000, 27000, 28000, 29000, 30000, 31000, 32000, 33000, 34000, 35000, 36000, 37000, 38000, 39000, 40000, 41000, 42000, 43000, 44000, 45000, 46000, 47000, 48000, 49000, 50000, 51000, 52000, 53000, 54000, 55000, 56000, 57000, 58000, 59000, 60000, 61000, 62000, 63000, 64000, 65000, 66000, 67000, 68000, 69000, 70000, 71000, 72000, 73000, 74000, 75000, 76000, 77000, 78000, 79000, 80000, 81000, 82000, 83000, 84000, 85000, 86000, 87000, 88000, 89000, 90000, 91000, 92000, 93000, 94000, 95000, 96000, 97000, 98000, 99000, 100000, 101000, 102000, 103000, 104000, 105000, 106000, 107000, 108000, 109000, 110000, 111000, 112000, 113000, 114000, 115000, 116000, 117000, 118000, 119000, 120000, or 121000. More specifically, it is expected that promoted grain refinement can be achieved in direct chill casting when using a jet having a Reynolds number at or above a threshold Reynolds number of at or approximately 14000, 15000, 16000, 17000, 18000, 19000, 20000, 21000, 22000, 23000, 24000, 25000, 26000, 27000, 28000, 29000, 30000, 31000, 32000, 33000, 34000, 35000, 36000, 37000, 38000, 39000, 40000, 41000, 42000, 43000, 44000, 45000, 46000, 47000, 48000, 49000, 50000, 51000, 52000, 53000, 54000, 55000, 56000, 57000, 58000, 59000, 60000, 61000, 62000, 63000, or 64000. As described in further detail herein, promoted grain refinement can be achieved in a multi-nozzle direct chill casting system when the sum of the Reynolds numbers of the multiple jets from the multiple nozzles is at or above any of the aforementioned threshold Reynolds numbers.

The foregoing description of the embodiments, including illustrated embodiments, has been presented only for the purpose of illustration and description and is not intended to be exhaustive or limiting to the precise forms disclosed. Numerous modifications, adaptations, and uses thereof will be apparent to those skilled in the art.

As used below, any reference to a series of examples is to be understood as a reference to each of those examples disjunctively (e.g., “Examples 1-4” is to be understood as “Examples 1, 2, 3, or 4”).

Example 1 is a casting system, comprising: a feed tube couplable to a source of molten metal; and a nozzle located at a distal end of the feed tube, the nozzle submersible in a molten sump for delivering the molten metal to the molten sump, wherein the nozzle is designed to supply the molten metal at a Reynolds number of at least 14,000.

Example 2 is the casting system of example(s) 1, wherein the nozzle is designed to supply the molten metal at a Reynolds number of at least 64,000,

Example 3 is the casting system of example(s) 1, wherein the nozzle is designed to supply the molten metal at a Reynolds number of at least 15000, 16000, 17000, 18000, 19000, 20000, 21000, 22000, 23000, 24000, 25000, 26000, 27000, 28000, 29000, 30000, 31000, 32000, 33000, 34000, 35000, 36000, 37000, 38000, 39000, 40000, 41000, 42000, 43000, 44000, 45000, 46000, 47000, 48000, 49000, 50000, 51000, 52000, 53000, 54000, 55000, 56000, 57000, 58000, 59000, 60000, 61000, 62000, or 63000.

Example 4 is the casting system of example(s) 1-3, further comprising a mold for receiving the molten metal, wherein the mold comprises one or more mold walls and a bottom block lowerable to support a solidifying ingot.

Example 5 is a casting system, comprising: at least one feed tube couplable to a source of molten metal; and a set of nozzles comprising one or more nozzles, wherein each of the one or more nozzles is located at a distal end of the at least one feed tube and is submersible in a molten sump for delivering the molten metal to the molten sump, wherein each of the one or more nozzles has an opening sized to achieve a jet of molten metal within the molten sump at a Reynolds number, wherein a sum of the Reynolds numbers of each jet of molten metal is at least 14,000.

Example 6 is the casting system of example(s) 5, wherein the openings of the one or more nozzles are sized such that the sum of the Reynolds numbers of each jet of molten metal is at least 64,000,

Example 7 is the casting system of example(s) 5, wherein the openings of the one or more nozzles are sized such that the sum of the Reynolds numbers of each jet of molten metal is at least 15000, 16000, 17000, 18000, 19000, 20000, 21000, 22000, 23000, 24000, 25000, 26000, 27000, 28000, 29000, 30000, 31000, 32000, 33000, 34000, 35000, 36000, 37000, 38000, 39000, 40000, 41000, 42000, 43000, 44000, 45000, 46000, 47000, 48000, 49000, 50000, 51000, 52000, 53000, 54000, 55000, 56000, 57000, 58000, 59000, 60000, 61000, 62000, or 63000.

Example 8 is the casting system of example(s) 5-7, wherein the set of nozzles includes at least two nozzles.

Example 9 is the casting system of example(s) 5-8, further comprising a mold for receiving the molten metal, wherein the mold comprises one or more mold walls and a bottom block lowerable to support a solidifying ingot.

Example 10 is a method, comprising: delivering molten metal from a metal source to a metal sump through a feed tube, wherein the molten metal generates one or more jets of molten metal within the metal sump upon exiting the feed tube, wherein each of the one or more jets of molten metal has a Reynolds number, and wherein a sum of the Reynolds numbers of each of the one or more jets of molten metal is at least 14,000.

Example 11 is the method of example(s) 10, wherein the sum of the Reynolds numbers of each of the one or more jets of molten metal is at least 64,000,

Example 12 is the method of example(s) 10, wherein the sum of the Reynolds numbers of each of the one or more jets of molten metal is at least 15000, 16000, 17000, 18000, 19000, 20000, 21000, 22000, 23000, 24000, 25000, 26000, 27000, 28000, 29000, 30000, 31000, 32000, 33000, 34000, 35000, 36000, 37000, 38000, 39000, 40000, 41000, 42000, 43000, 44000, 45000, 46000, 47000, 48000, 49000, 50000, 51000, 52000, 53000, 54000, 55000, 56000, 57000, 58000, 59000, 60000, 61000, 62000, or 63000.

Example 13 is the method of example(s) 10-12, further comprising solidifying the molten metal into an ingot using a mold comprising one or more mold walls and a bottom block lowerable to support the solidifying ingot.

Example 14 is a metal product cast according to the method of example(s) 10-13.

Example 15 is the metal product of example(s) 14, wherein an average grain size of the metal product is at or below approximately 130 μm.

Example 16 is the metal product of example(s) 14, wherein an average grain size of the metal product is at or below approximately 129 μm, 128 μm, 127 μm, 126 μm, 125 μm, 124 μm, 123 μm, 122 μm, 121 μm, 120 μm, 119 μm, 118 μm, 117 μm, 116 μm, 115 μm, 114 μm, 113 μm, 112 μm, 111 μm, 110 μm, 109 μm, 108 μm, 107 μm, 106 μm, 105 μm, 104 μm, 103 μm, 102 μm, 101 μm, or 100 μm.

Example 17 is the metal product of example(s) 14-16, wherein a maximum grain size of the metal product is at or below approximately 250 μm.

Example 18 is the metal product of example(s) 14-16, wherein a maximum grain size of the metal product is at or below approximately 245 μm, 240 μm, 235 μm, 230 μm, 225 μm, 220 μm, 215 μm, 210 μm, 205 μm, 200 μm, 195 μm, 190 μm, 185 μm, 180 μm, 175 μm, 170 μm, 165 μm, 160 μm, 155 μm, 150 μm, 145 μm, 140 μm, 135 μm, or 130 μm.

Example 19 is the metal product of example(s) 14-18, wherein a grain size spread of the metal product is at or below approximately 130 μm.

Example 20 is the metal product of example(s) 14-18, wherein a grain size spread of the metal product is at or below approximately 125 μm, 120 μm, 115 μm, 110 μm, 105 μm, 100 μm, 95 μm, 90 μm, 85 μm, 80 μm, 75 μm, 70 μm, 65 μm, 60 μm, or 55 μm.

SHEAR INDUCED GRAIN REFINEMENT OF A CAST INGOT

Information

Publication Number

Date Filed

Date Published

Inventors

Original Assignees

CPC

International Classifications

Abstract

Description

Claims

CROSS REFERENCE TO RELATED APPLICATIONS

Provisional Applications (1)