Materials Transactions, Vol. 43, No. 1 (2002) pp. 24 to 29c©2002 The Japan Institute of Metals

Continuous Semi-Solid Casting Process for Aluminum Alloy Billets

Hakaru Nakato1, Michio Oka2, Seiji Itoyama3, Masao Urata1,Tatsuo Kawasaki1, Ko-ichi Hashiguchi1 and Shinobu Okano4

1Process & Products Engineering Division, Kawasaki Steel Techno-research Corp., Chiba 260-0835, Japan2Chiba Works, Kawatetsu Machinery Co., Ltd., Chiba 260-0835, Japan3Research Laboratories, Kawasaki Steel Corp., Chiba 260-0835, Japan4Rheo-technology, Ltd., Chiba 260-0835, Japan

A new process for continuous semi-solid casting of billets of aluminum alloys was developed. Round aluminum alloy billets 75 mm and150 mm in diameter are continuously cast in a semi-solid state by agitating the alloy in the agitating vessel with a mechanical screw and/or anelectromagnetic stirrer. The solidification structure of the billets obtained by this process is a mixed structure of granular particles and a fineeutectic structure, except in a thin chill layer about 2 mm in thickness, which shows a dendrite structure. It was possible to use billets of AC4Calloy obtained by this process in the thixoforming process without surface conditioning.

(Received September 17, 2001; Accepted November 19, 2001)

Keywords: aluminium alloy, rheo-casting, semi-solid casting, continuous casting, thixoforming, round billet

1. Introduction

Products manufactured by thixoforming of aluminum al-loys cast from reheated semi-solid billets show good sound-ness and excellent mechanical properties.

Since thixoforming was first proposed by Flemings et al.1)

in 1972, many efforts have been made to realize a practicalthixoforming process,2–9) and theoretical investigation10, 11) ofthis type of processing has also been reported.

Bertrand and Patrick2) studied the development of a con-tinuous semi-solid casting process for 7075 aluminum alloyusing electromagnetic stirring in the mold. Idegomori et al.,5)

reported the results of the application of thixotropic technol-ogy to automobile parts using billets cast with electromag-netic stirring in the mold. Products manufactured by this tech-nology have been applied to outer rigger of the rear part ofautomobiles.

Uetani et al.,6) developed a process for semi-solid cast 7075aluminum alloy billets using a mechanical stirring method.The tensile strength of the product after T6-treatment wasnearly equal to that of hot extruding products, but elongationwas somewhat lower than with hot extruding.

Although some studies have reported the production ofsemi-solid billets for use in thixoforming of aluminum alloys,the current production capacity is inadequate to supply themany different kinds of billets required by customers imme-diately.

Hence, semi-solid materials are strongly desired for thethixoforming process for manufacturing sound products in thealuminum industry.

Cooperative research by Rheo-technology, Ltd., KawasakiSteel Techno-research Corp., Kawatetsu Machinery Co., Ltd.,and two billet users successfully realized a new continuoussemi-solid casting process for aluminum alloys. This paperpresents an outline of the process and the operating results.

2. Experimental Procedure

2.1 Experimental apparatusThe appearance of the continuous semi-solid casting pro-

cess for billets of aluminum alloys is shown in the photographin Fig. 1. The main specifications of the process are given inTable 1. The process is characterized by two main parts, an

Fig. 1 Appearance of continuous semi-solid casting machine for aluminumalloy billets.

Table 1 Main specifications of continuous semi-solid caster.

Melting furnace Capacity (kg) 200

Heater Propane burner

Agitating vessel Screw speed (min−1) Max. 800

Solid fraction 0–0.4

Billet size Diameter (mm) 75, 150

Length (m) Max. 4

Pinch roll Withdrawal speed (m/min) 0.05–2.0

equipment Drive system AC motor

Continuous Semi-Solid Casting Process for Aluminum Alloy Billets 25

Fig. 2 Schematic diagram of major parts of continuous semi-solid castingprocess.

agitating vessel, which is used to produce semi-solid mate-rial (material with coexistent solid and liquid phases) and acooling mold for producing the solidified shell (billet), whichare independent parts of the process. As shown in Fig. 2, theprocess equipment comprises eight units: a melting furnace,a tundish or teeming gutter, a rotating shaft (screw), an elec-tromagnetic coil, an agitating vessel, a feed nozzle and breakring, a cooling mold, and pinch rolls. A general descriptionof this equipment is presented below.(1) Melting furnace

Maximum capacity of 200 kg; the alloy is heated by apropane burner from outside the graphite crucible.(2) Tundish or teeming gutter

The wall of this equipment is coated with a refractory mor-tar and is preheated from the top with a propane burner. Thepreheating system in the tundish or teeming gutter and the ag-itating screw comprises a pair of propane burners.(3), (4) Rotating shaft, Electromagnetic coil

A mechanical screw installed in the vessel was also usedand/or an electromagnetic stirrer was applied to the alloy inthe vessel from outside the wall (75 mm round billet).(5) Agitating vessel

The wall, which is made of stainless steel, is water-cooled.A mechanical screw installed in the vessel was used (150 mmround billet). A silica crucible was used and was preheatedelectrically by a spiral rod heater installed in the crucible. Themechanical screw installed in the vessel with a rotating shaftwas also used and/or an electromagnetic stirrer was appliedto the alloy in the vessel from outside the wall (75 mm roundbillet).(6) Feed nozzle and break ring

A feed nozzle made of stainless steel was used with150 mm round billets, and a feed nozzle made of graphite andbreak ring made of graphite or silica were used with 75 mmround billets. The feeding area between the agitating vesseland the mold was electrically heated using a heating coil fromoutside the feed nozzle and break ring to prevent temperaturedrop and maintain a constant temperature in these parts.(7) Mold

Table 2 Experimental conditions.

(150 mm round billet)

Al alloys used AC4C, ADC12

Agitating vessel Material Stainless steel

Mold Material Al alloy

Length (mm) 150

(75 mm round billet)

Al alloys used AC4C, AC1B

Agitating vessel Material Silica

Feed nozzle and break Silica, graphite

ring materials

Mold Lining materials Al alloy, graphite

Length (mm) 60, 100, 150

Conditions of

electromagnetic

stirrer

Intensity at mold center (T) Max. 0.3

Applied current (A) 210, 430

Applied frequency (Hz) 5, 10

Two types of molds (75 mm and 150 mm in diameter) wereused. The molds were made of an aluminum alloy witha water-cooling channel. The mold length was from 60 to150 mm (100 mm in most cases).(8) Pinch roll equipment

A 2-high pinch roll device driven by an AC motor was used.Cast billets were held by an oil hydraulic mechanism.

2.2 Experimental conditionsThe experimental conditions are listed in Table 2. In both

campaigns in these experiments (i.e., 150 mm and 75 mmround billets), the main material used in casting was AC4Calloy.

3. Results and Discussion

3.1 Casting practiceThe casting conditions used in the experiment are listed in

Table 3. Temperature control of the semi-solid aluminum al-loy in the agitating vessel was a key factor in successful cast-ing.〈Casting practice for 150 mm round billets〉

Heat loss in the agitating vessel was mainly controlled bythe heat loss through the water-cooled stainless steel wall.Therefore, the quantity of cooling water was strictly regulatedto control the condition of heat flow from the bulk alloy to thecooling water.

By applying a cooling water flow rate of less than 100 l/minfrom outside the stainless steel wall, mild cooling of the alu-minum alloy in the agitating vessel was effectively accom-plished, enabling stable casting of billets.

The relationship between the motor torque of the rotatingscrew and the solid fraction of the melt is shown in Fig. 3.The solid fraction was calculated from the temperature in thesemi-solid region measured 30 mm below the tip of the screw.At solid fractions of more than 0.3, motor torque increased,and when the solid faction exceeded 0.4, large torque fluctua-tions were also observed.

26 H. Nakato et al.

Table 3 Casting conditions used in experiments.

(150 mm round billet)

Metal temperature in tundish (K) 953 (ADC12)

993–1013 (AC4C)

Teeming rate (kg/min) 100–150

Holding time of metal in mold after 30

pouring (s)

Withdrawal speed (m/min) Initial stage 0.2–0.3

Steady state 0.2–0.5

Cooling water intensity in agitating 20–200

vessel (l/min)

Cooling water intensity in mold (l/min) 800

Heating temperature of feeding nozzle (K) 923–953

Agitating screw speed (min−1) Initial stage 100

Steady stage 600

Setting torque limit of agitating screw (J) 167

(75 mm round billet)

Metal temperature in furnace (K) 908–923

Preheating temperature of outer surface of Upper 503–593crucible (K) Middle 553–713

Lower 483–653

Preheating temperature of feed nozzle (K) 883–903

Preheating temperature of break ring (K) 868–893

Teeming rate (kg/min) 150

Cooling water intensity in mold (l/min) 100–450

Initial holding time of teemed alloy in mold (s) 15–25

Agitating screw speed (min−1) Initial stage 10–100

Steady state 30–400

Withdrawal speed (m/min) Initial stage 0.15–0.20

Steady state 0.15–0.40

Fig. 3 Relationship between motor torque of agitating screw and solid frac-tion calculated from temperature measured 30 mm below tip of screw.

The relationship between the solid fraction and the temper-ature of the AC4C alloy calculated thermodynamically by us-ing “THERMOCALC” software is shown in Fig. 4. In Fig. 4,the experimental results obtained by S. Okano are also plot-ted. In this paper, the data presented by Okano were used toobtain the relationship between the solid fraction and temper-

Fig. 4 Relationship between temperature and solid fraction of AC4C alloy.

Fig. 5 Appearance of cast billet of AC4C alloy.

ature. The slope of the relationship between the solid fractionand temperature of the AC4C alloy in the vessel is steeperwith solid fractions of more than 0.6 than with smaller solidfractions. When the temperature drops less than 850 K, largefluctuation in the solid fraction arises from small fluctuationsin temperature. In high solid fraction region exceeds 0.4, evensmall temperature deviations cause large fluctuations in thesolid fraction, making stable casting impossible.〈Casting practice for 75 mm round billets〉

Heat flow from the agitating vessel was mainly controlledby the heat loss through the rotating screw rod. Therefore, thetemperature of the rotating screw rod was preheated strictlyby controlling the preheating conditions. With AC4C alloybillets 75 mm in diameter, stable casting of sound billets withgood surface quality was achieved by optimizing the castingconditions.

An example of the surface appearance of the cast billet isshown in the photograph in Fig. 5. Billets of AC4C alloyobtained by the newly developed process could be cast suc-cessfully by thixoforming without surface conditioning.

Two typical examples of the measured change in tempera-ture in these experiments are shown in Fig. 6. In this figure,run No. 18 was successfully cast, but run No. 20 ended infailure. The temperature in the break ring installed betweenthe feed nozzle and the mold was measured by a thermocou-

Continuous Semi-Solid Casting Process for Aluminum Alloy Billets 27

Fig. 6 Two typical examples of temperature change in experiments.

Fig. 7 Change over time in temperature at centerline of billets (run/moldlength: run No. 35/60 mm, run No. 36/100 mm).

Fig. 8 Change in mold heat flux during continuous semi-solid casting ofAC4C alloy.

ple embedded at a depth of 3 mm from the inner surface. Inthe failed run No. 20, large temperature fluctuations were ob-served. This indicated that the solidified shell was sticking tothe surface of the break ring and then breaking away periodi-cally.

The change over time in the temperature at the centerlineof an AC4C alloy billet is shown in Fig. 7. Although thetemperature was substantially constant as long as the billetwas in the mold, the temperature dropped rapidly at the end ofthe mold due to a cooling-water film flowing from the loweroutlet of the mold.

The change in heat flux in the mold during continuoussemi-solid casting of AC4C alloy is shown in Fig. 8. Theheat flux in the upper and lower parts of the mold was mea-sured at points of 30 mm and 60 mm from the top of the mold,

respectively. The heat flux in the upper part of the mold wasapproximately 400×104 (W·m−2·K−1), whereas that in lowerpart was approximately 250 × 104 (W·m−2·K−1), with smallfluctuations over time.

3.2 Quality of billets3.2.1 Solidification structure

The solidification structure of an aluminum alloy (AC4C)billet obtained by this continuous semi-solid casting processis shown in Fig. 9.

With 150 mm round billets, as shown in the upper part ofFig. 9, a mixed structure consisting of granular particles and afine eutectic structure was observed, except in the chill layerwhich formed 2 to 3 mm from the surface and showed a den-dritic structure. The average size of the granular particles was75 µm, and the ratio of granular particles was in the range of0.61 to 0.69 in the billet cross section. The observed values of0.61 to 0.69 were much greater than the values of 0.2 to 0.4which had been calculated from the temperature measured at30 mm below the tip of the screw in the semi-solid region.This indicates that the primary granular structure grew in themold during solidification.

As shown in the lower part of Fig. 9, a mixed structuresimilar to that in the upper part of Fig. 9 was observed with75 mm round billets except in the chill layer. The granularstructure somewhat changed from the globular structure thatexisted in the initial period of casting. An example of a tree-ring-like (coring structure) in the granular structure associatedwith the solidification process is shown in Fig. 10. The rela-tionship between the solid fraction calculated from the tem-perature measured 30 mm below the tip of the screw in thesemi-solid region and the solidification structure of the solid-ified billet is shown in Fig. 11. In the range of solid fractionsfrom 0 to 0.2, rosette-type crystals formed. However, at solidfractions of more than 0.2, granular crystals formed.

In the experiment, the effect of rotating method on the so-lidification structure in the billet was not clarified.3.2.2 Mechanical properties of as-cast billets at elevated

temperaturesThe high temperature deformation behavior of as-cast

billets was examined using a hot forging test machine(“FORMASTER”). The size of the test pieces prepared fromthe as-cast billet was 8 mm in diameter and 12 mm in height.The effect of the test temperature on the maximum deforma-tion resistance is shown in Fig. 12. The maximum deforma-tion resistance depended on the billet size. At a test tempera-ture of 773 K, the maximum deformation resistance of a billetwith a diameter of 75 mm was greater than that of a billet witha diameter of 150 mm due to the difference in the solidifiedstructure.

Moreover, above 823 K, the maximum deformation resis-tance decreased abruptly as the test temperature increased,being approximately 30 MPa at 828 K, but falling to below10 MPa at 838 K. Above 833 K, the maximum deformationresistance of billets cast in the semi-solid state was lower thanthat of conventional billets cast with a superheat in the moltenmaterial, and at 838 K, the difference in the maximum de-formation resistance of the two types of billets became morepronounced.

28 H. Nakato et al.

Fig. 9 Cross-sectional view of solidification structures of 150 mm round billet (upper part of figure; estimated solid fraction, 0.30 atpoint 30 mm below tip of screw) and of 75 mm round billet (lower part; estimated solid fraction, 0.14). Percentage values are thecross-sectional area ratio of primary granules.

Fig. 10 Coring structure formed in the granular structure associated withsolidification.

4. Conclusion

A new process for continuous semi-solid casting of billetsof aluminum alloys was developed. The results obtained inexperimental campaigns with two different sizes are summa-rized below.

(1) Round billets of aluminum alloys with diameters of75 mm and 150 mm were continuously cast in the semi-solidstate by agitating the alloy in the agitating vessel with a me-chanical screw and/or an electromagnetic stirrer.

(2) The solidification structure of the billets obtained bythis process was a mixed structure consisting of granular par-ticles and a fine eutectic structure, except in a thin chill layer(about 2 mm), which showed a dendrite structure.

(3) With the 150 mm round billets, the average size ofthe granular particles was 75 µm, and the ratio of granularparticles was in the range of 0.61 to 0.69 in the billet crosssection. With the 75 mm round billets, when the solid fractioncalculated from the temperature measured 30 mm below thetip of the screw was in the range of 0 to 0.2, a rosette-typestructure formed. On the other hand, when the solid fractionexceeded 0.2, the granular crystal structure formed.

(4) The maximum deformation resistance at high temper-ature depended on the billet size. At a test temperature of773 K, the maximum deformation resistance of the 75 mmφ

billets was greater than that of the 150 mmφ billets due to thedifference in the solidified structure. Above 823 K, the max-imum deformation resistance abruptly decreased as the testtemperature increased. Deformation resistance was approxi-mately 30 MPa at 828 K, but dropped to less than 10 MPa at838 K. Above 833 K, the maximum deformation resistance ofbillets cast in the semi-solid state was lower than that of con-ventional billets cast with a superheat in the molten metal. At838 K, the difference in the maximum deformation resistanceof the two types of billets became more pronounced.

(5) It was possible to use billets of AC4C alloy obtainedby this process in the thixoforming process without surfaceconditioning.

Continuous Semi-Solid Casting Process for Aluminum Alloy Billets 29

Fig. 11 Effect of solid fraction calculated from temperature measured 30 mm below tip of screw on solidification structure of billet.

Fig. 12 Effect of test temperature on maximum deformation resis-tance(150D: 150 mm billet, die cast; 150S: 150 mm billet, semi-solid cast;75D: 75 mm billet, die cast; 75S: 75 mm billet, semi-solid cast).

Acknowledgements

The authors wish to express their sincere thanks to theJapan Small and Medium Enterprise Corp., which sponsoredthis work. The authors are also particularly grateful tomembers of the technical committee of the project for helpfuldiscussions.

REFERENCES

1) D. B. Spencer, R. Mehrabian and F. C. Flemings: Metall. Trans. 3 (1972)1925–1930.

2) C. Bertrand and P. Patrick: Proc. of the 4th Int. Conference on

Semi-Solid Processing of Alloys and Composites, (The University ofSheffield, England, June, 1996) pp. 169–173.

3) S. Blais, W. Loue and C. Pluchon: Proc. of the 4th Int. Conference onthe Semi-Solid Processing of Alloys and Composites, (The Universityof Sheffield, England, June 19–21, 1996) pp. 187–192.

4) D. Bernhard, H. M. Spath and P. R. Sahm: Proc. of the 5th Int. Confer-ence on the Semi-solid Processing of Alloys and Composites, (Golden,Colorado, USA, June 23–25, 1998) pp. 51–55.

5) T. Idegomori, K. Sakamoto, H. Hirono, E. Masuda and N. Saito: GPC’99 New Materials & Development Processes (1999) pp. 20–28.

6) Y. Uetani, H. Takagi, K. Matsuda and S. Ikeno: Materia Japan 39 (2000)769–771.

7) T. Motegi, N. Ogawa, K. Kondo, C. Liu and S. Aoyama: Proc. ofICAA-6, (1998) pp. 297–301.

8) “Semi-Solid Forging at Alumax”: Light Metal Age, (1994) Oct.,pp. 32–34.

9) B. C. Pai, R. M. Pillai and K. G. Satyanarayana: Indian Foundry J. 45(1999) 47–57.

10) N. S. Kim and C. G. Kang: J., Materials Processing Tech. 103 (2000)237–246.

11) A. N. Alexandrou, G. R. Burgos and V. M. Entov: Mater. Process Com-put. Age 3 (2000) pp. 161–170.