Effect of the salt addition practice on the grain refining efficiency of Al–Ti–B master alloys

Effect of the salt addition practice on the grain refining efficiency of Al–Ti–B master alloys

Journal of Alloys and Compounds 420 (2006) 207–212 Effect of the salt addition practice on the grain refining efficiency of Al–Ti–B master alloys Y¨u...

898KB Sizes 0 Downloads 0 Views

Journal of Alloys and Compounds 420 (2006) 207–212

Effect of the salt addition practice on the grain refining efficiency of Al–Ti–B master alloys Y¨ucel Birol ∗ ˙ Materials Institute, Marmara Research Center, TUBITAK, Gebze, Kocaeli, Turkey Received 18 October 2005; received in revised form 1 November 2005; accepted 3 November 2005 Available online 15 December 2005

Abstract The impact of the salt addition practice on the microstructure and grain refining efficiency of Al–Ti–B alloys produced by the “halide salt” route was investigated. The grain refining performance of an experimental Al–5Ti–1B master alloy was optimized when the halide salts were pre-mixed before addition to aluminium melt at 800 ◦ C during the production of the grain refiner. The stirring action provided during salt addition was found to degrade, while a high rate of addition was found to improve, the grain refining efficiency. In view of the above, an improved salt addition practice to ensure an exceptional grain refining performance is claimed to comprise the following steps: melting commercial purity aluminium ingot; addition of pre-mixed salts to molten aluminium at 800 ◦ C, at once to facilitate a rapid salt reaction, gently mixing the salts with the aluminium melt without introducing any stirring. The grain refiner master alloy thus produced gives an average grain size of 102 ␮m 2 min after inoculation. © 2005 Elsevier B.V. All rights reserved. Keywords: Metals; Casting; Al–Ti–B master alloy; Grain refinement

1. Introduction Addition of master alloys to molten aluminium produces fine, equiaxed grains after solidification which otherwise tend to be coarse and columnar [1–5]. A fine, equiaxed grain structure imparts to a casting, high toughness, high yield strength, excellent formability, good surface finish and improved machinability [6,7]. Furthermore, a sound grain-refining practice avoids hot tearing, allows a marked increase in casting speed and improves the homogeneity of the cast structure by refining the distribution of secondary phases and microporosity [8]. The use of grainrefining master alloys in casting of ingots, billets and strip, has thus become a standard practice in aluminium foundries worldwide. The grain refinement of aluminium alloys by melt inoculation is generally achieved by introducing into the melt, Al–Ti–B master alloys which typically consist of TiAl3 and TiB2 particles in an aluminium matrix [7,9]. Commercial production of these alloys involves the addition of K2 TiF6 and KBF4 salts to molten aluminium above 700 ◦ C [10–12]. These complex salts react with liquid aluminium quickly and very efficiently pro∗

Tel.: +90 262 6412300x3481; fax: +90 262 6412309. E-mail address: [email protected]

0925-8388/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jallcom.2005.11.010

ducing a melt with dispersed particles of TiAl3 and (Al, Ti)B2 and high yields of Ti and B in the final alloy. In spite of a large volume of work in the field of aluminium grain refinement [13–16], the production of master alloys themselves has not receieved much attention. The microstructure and the performance of a grain refiner, on the other hand, are known to be highly sensitive to the processing parameters used in the production of master alloys [17–19]. The reaction temperature and time were found to have a strong impact on the microstructure, the morphology of TiAl3 particles in particular, of the Al–Ti–B master alloys, dictating their grain refining efficiency [18]. Consequently, the results obtained in regard to grain refinement with Al–Ti–B alloys differ appreciably. The effect of holding temperature, holding time, and stirring conditions during holding on the microstructure and grain refinement performance of Al–Ti–B master alloys were recently identified [20]. The present work was undertaken to investigate the impact of the salt addition practice on the microstructure and grain refining efficiency of Al–Ti–B alloys. 2. Experimental A series of experimental Al–5Ti–1B master alloys were produced in the laboratory by reacting K2 TiF6 and KBF4 salts with molten aluminium. The starting materials were aluminium (99.7% Al), containing 0.001% Ti and <0.0002% B,

208

Y. Birol / Journal of Alloys and Compounds 420 (2006) 207–212

Table 1 Effect of salt addition parameters on the Ti recovery in the final grain refiner alloy Alloy no.

Addition sequence

Addition temperature (◦ C)

Addition rate

Stirring at addition

Ti pick-up (%)

Grain size 2 min after inoculation (␮m)

1 2 3 4 5 6 7 8 9 10

KBF4 first K2 TiF6 first Pre-mixed Mixed + melted Pre-mixed Pre-mixed Pre-mixed Pre-mixed Pre-mixed Pre-mixed

800 800 800 800 750 850 900 800 850 800

At once At once At once At once At once At once At once Gradual Gradual At once

No stirring No stirring No stirring No stirring No stirring No stirring No stirring No stirring No stirring Mechanical

83.6 94.8 99.4 90.6 98.9 98.7 97.1 93.4 93.2 95.7

290 218 102 136 128 161 180 184 240 162

and K2 TiF6 and KBF4 salts of commercial purity. Production campaigns were carried out on a 1.5 kg batch scale. Aluminium metal was brought to reaction temperature using an electrical resistance furnace. The temperature of the melt at the time of salt addition (reaction temperature) ranged from 750 to 900 ◦ C in different experiments. The lower limit of 750 ◦ C was a practical limit to maintain the molten alloy as a castable liquid while 900 ◦ C was identified as the upper limit to reduce processing costs and to avoid excessive oxidation of the melt. The fluoride salts, weighted in proportions corresponding to a Ti/B ratio of 5, were added to the melt either at once or gradually over a period of time, either premixed or seperately, one after the other. The pre-mixed salts were melted before addition in one of these experiments. Different practices, ranging from gentle mixing to vigorous mechanical stirring, were employed during the addition of salts. The temperature of the melt started to increase when the reaction between the aluminium melt and the fluoride salts was underway, immediately after salt addition. The melt was held at 800 ◦ C for 30 min in an electric resistance furnace once the temperature of the melt ceased to increase, under conditions which were optimized in a recent study [20]. A total of 10 experimental alloys were prepared using different salt addition sequences, addition temperatures and addition rates, with and without stirring during addition (Table 1). The experimental master alloys thus produced were assessed for their grain refining performances. Aluminium ingot with a purity of 99.7% Al, weigthing 1000 g, was melted in a medium frequency induction furnace and its temperature was brought to 710 ◦ C. A reference sample was taken from the melt in order to identify the grain size before inoculation. Four grams of the experimetal alloy was then added to the melt. This is the exact amount of an Al–5Ti–1B master alloy needed to bring the Ti concentration of 1 kg aluminium melt to 0.02%. The melt was stirred with a graphite rod for 30 s right after inoculation. Samples were taken from the melt 2, 5, 15, 30 and 60 min after the addition and were solidified in small copper molds with a diameter of 25 mm and a height of 50 mm. Measures were taken to keep the temperature of the melt within 710 ± 5 ◦ C during the entire process. The samples thus produced were sectioned 20 mm from the bottom surface. Standard metallographic procedures were employed to prepare these sections for grain size measurements. They were etched with Poulton’s reagent and were then examined under a light microscope. Additionally, the 2-min samples in the present work were anodized in Barker’s solution to facilitate grain size measurements with the line intercept method.

3. Results and discussion 3.1. Salt addition sequence Ti pick-up in alloy 1, produced by adding into the aluminum melt, the KBF4 salt first, was as low as 83.6%. There was a marked increase in Ti pick-up when K2 TiF6 , instead of KBF4 , was added to molten aluminium first (Table 1). This contrasts the finding of Murty et al. [21], who measured a higher Ti pickup when the KBF4 salt was added first. The Ti recovery was the highest (99.4%) when the two salts were pre-mixed before

addition (alloy 3), but dropped to 90.6% when the pre-mixed salts were added in molten state (alloy 4). It is fair to conclude that pre-mixing the K2 TiF6 and KBF4 salts prior to addition helps to achieve a nearly complete salt reaction and high Ti recoveries. Alloys 1–4 all revealed predominantly blocky TiAl3 particles (Fig. 1a–d). Aluminide particles were relatively larger in alloy 2, possibly due to the earlier nucleation of TiAl3 particles (Fig. 1b). Alloy 3, for which the two salts were pre-mixed before addition, revealed a more uniform and balanced dispersion of aluminide and boride particles (Fig. 1c). It is interesting to note that the population of TiB2 particles appeared to be the lowest in alloy 1 which was produced by adding to the melt, the KBF4 salt first (Fig. 1a). The salt addition sequence appeared to have a big impact on the grain refining performance of the Al–5Ti–1B master alloy (Fig. 2). Poor results with a mixture of columnar and equiaxed grains were obtained when the KBF4 salt was added to the melt first, during the production of the grain refiner (Fig. 2a). Addition of the K2 TiF6 salt first instead, has improved the grain refining performance (Fig. 2b) which was remarkable when the salts were pre-mixed before addition (Fig. 2c). Inoculation with alloy 3 has produced a fine equiaxed grain structure across the entire section of the sample which was more or less retained until 60 min after inoculation. The average grain size 2 min after inoculation with alloys 1–3 were 290, 218 and 102 ␮m, respectively, implying two to three-fold decrease in grain size with the latter alloy. The very impressive grain refining efficiency of alloy 3 can be linked with the higher Ti content and the uniform and balanced dispersion of aluminide and boride particles in this alloy. A slight deterioration in the grain refining performance was noted when the salt mixture was added to the aluminium melt in the molten state (the average grain size 2 min after inoculation was 136 ␮m), possibly due to the relatively lower Ti content of this alloy. In view of the above, experimental master alloys for further studies have been prepared by pre-mixing the two salts before addition to molten aluminium. 3.2. Salt addition temperature The pre-mixed salts were added to molten aluminium at several temperatures between 750 and 900 ◦ C. The rest of the

Y. Birol / Journal of Alloys and Compounds 420 (2006) 207–212

209

Fig. 1. The effect of salt addition parameters on the microstructure of Al–5Ti–1B master alloys: (a) alloy 1, (b) alloy 2, (c) alloy 3, (d) alloy 4, (e) alloy 5, (f) alloy 6, (g) alloy 7, (h) alloy 8, (i) alloy 9 and (j) alloy 10.

210

Y. Birol / Journal of Alloys and Compounds 420 (2006) 207–212

Fig. 2. The effect of salt addition sequence on the grain refining performance of the master alloy: (a) KBF4 salt (alloy 1), (b) K2 TiF6 salt added first (alloy 2), (c) salts are pre-mixed before addition (alloy 3) and (d) salts are pre-mixed and melted before addition (alloy 4).

production cycle involved holding of the molten alloy at 800 ◦ C for 30 min in an electric resistance furnace without stirring until casting. The Ti pick-up in alloys 3, 5–7, thus obtained, were very similar and were all above 97%, suggesting that the addition temperature had little effect, if any, on the Ti recovery (Table 1). The microstructural features of these alloys were also found to be very similar (Fig. 1c, e–g). TiAl3 particles once again were of the blocky variety and were generally smaller than 20 ␮m. Several non-blocky aluminide particles were noted when the salt mixture was added to molten aluminium at 900 ◦ C. The TiB2 particles, on the other hand, were generally smaller than 1 ␮m at all addition temperatures employed. The grain refinement performances of the experimental Al–5Ti–1B alloys produced by salt addition between 750 and 900 ◦ C are illustrated in Fig. 3. Complete conversion of coarse columnar grains to fine equiaxed ones occurred within 2 min of inoculation with these alloys. While their performances were all adequate and appeared to be similar to the naked eye, investigation of the 2-min samples anodically etched with Barker’s solution demonstrated that they are indeed different. The average grain sizes 2 min after inoculation with alloys produced by salt addition at 750, 800, 850 and 900 ◦ C (alloys 5, 3, 6 and 7) were 128, 102, 161 and 180 ␮m, respectively. Alloy 3, produced by the addition of pre-mixed salts at 800 ◦ C, once again, stands out with a remarkable grain refinement performance, suggesting that the grain refining efficiency is optimized at a salt addition temperature of 800 ◦ C. The degradation in the grain refinement performance when the salt addition was performed above 850 ◦ C, is claimed to be due to a decreasing number of blocky aluminide particles, which are reported to be more effective in grain refinement [18].

production of alloy 8, by adding the salt mixture to the melt gradually, over a period of time. As the reaction between the fluoride salts and the aluminium melt is strongly exothermic, the rate of salt addition is expected to affect the reaction step also temperature-wise. Hence, the gradual salt addition practice was repeated at a higher melt temperature (850 ◦ C) in order to compensate for the loss of melt heating in the case of gradual addition. The Ti pick-up, microstructural features and the grain refining performances of alloys 8 and 9, thus produced, were compared with those of alloy 3, which was produced by adding the salt mixture to the melt at 800 ◦ C, at once. The Ti pick-up in alloys 8 and 9, produced by adding the salt mixture into the melt gradually were similar yet lower than when it was added at once (93.4 and 93.2%, respectively, instead of 99.4%). The microstructural features of these alloys, on the other hand, were all alike with blocky TiAl3 particles ranging in size from several ␮m to approximately 20 ␮m (Fig. 1c, h, i). There was hardly any change in the population and the size of the aluminide and boride particles with salt addition rate. The inoculated grain sizes were markedly coarser, however, when the salt mixture was added to the aluminium melt gradually during the manufacture of the master alloy (Fig. 4). The average grain size 2 min after inoculation with alloy 8 was nearly twice as large as that obtained with alloy 3 (184 ␮m instead of 102 ␮m). Such a change in the grain refining efficiency cannot be accounted for by the relatively small decrease in the Ti recovery. The difference in the inoculated grain sizes were even larger, in favor of the higher rate of addition, when the addition temperature was increased to 850 ◦ C, in spite of very similar microstructural features observed in the two alloys. 3.4. Stirring during salt addition

3.3. Salt addition rate The reaction between the aluminium melt and the salt mixture was allowed to last longer (20 min instead of 2 min) in the

Ti pick-up in alloy 10 produced with mechanical stirring during salt addition dropped slightly to 95.7% (Table 1), possibly due to a more extensive oxidation as mechanical stirring pro-

Y. Birol / Journal of Alloys and Compounds 420 (2006) 207–212

211

Fig. 3. The grain refinement performances of the experimental Al–5Ti–1B master alloys produced with different reaction temperatures: (a) 750 ◦ C (alloy 5), (b) 800 ◦ C (alloy 3), (c) 850 ◦ C (alloy 6) and (d) 900 ◦ C (alloy 7).

Fig. 4. The grain refinement performances of the experimental alloys produced with different salt addition rates: (a) mixture added at once, within 2 min (alloy 3), (b) mixture added gradually in 20 min at a reaction temperature of 800 ◦ C (alloy 8) and (c) mixture added gradually in 20 min at 850 ◦ C (alloy 9).

moted atmospheric exposure. Microstructural features of this alloy were very similar with those of alloy 3 produced without stirring, with predominantly blocky aluminide particles and very fine boride particles as usual (Fig. 1c and j). The grain refining response of the two alloys, however, were different (Fig. 5). The Al–5Ti–1B alloy (alloy 10), stirred

mechanically in the molten state during salt addition has produced equiaxed grains across the section of the inoculated samples which were relatively coarser than those obtained in the case of alloy 3. The average grain size 2 min after inoculation with alloy 10 was 162 ␮m. It is, thus, fair to conclude that alloy 3, which was produced without stirring during salt addition at

Fig. 5. The grain refining performances of the two alloys, produced (a) by providing a mechanical stirring action during salt addition (alloy 10) and (b) without stirring (alloy 3).

212

Y. Birol / Journal of Alloys and Compounds 420 (2006) 207–212

800 ◦ C, was a better grain refiner than alloy 10. The mechanical stirring action introduced in the course of salt addition clearly had a detrimental effect on the grain refining efficiency of the Al–Ti–B master alloy, although not as intensive as that encountered in the case of stirring during holding [20]. Mixing of the potassium aluminium fluoride salt with the molten alloy, once again, is believed to be responsible for the loss of grain refining effect. The salt entrapped in the molten alloy takes part in the wetting of the boride particles, impairing their potency as nucleation sites and leads to their agglomeration at the same time, as demonstrated in [22].

Acknowledgements It is a pleasure to thank C. Kubilay, O. C¸akır and F. Alageyik of Marmara Research Center for their help in the experimental part of this work. References [1] [2] [3] [4]

4. Summary

[5]

Different salt addition practices produced a substantial change in the grain refining efficiencies of the experimental Al–5Ti–1B master alloys. Mixing the halide salts before addition and a high rate of addition both help to improve the grain refining efficiency while stirring action provided during addition and reaction temperatures exceeding 850 ◦ C work against an adequate grain refining performance. In view of the above, an improved salt addition practice to ensure an exceptional grain refining performance is claimed to comprise the following steps: melting commercial purity aluminium ingot; addition of pre-mixed salts to molten aluminium at 800 ◦ C, at once to facilitate a rapid salt reaction, gently mixing the salts with the aluminium melt without introducing any stirring. The grain refiner master alloys, further processed according to the practice outlined in [20], gives a very fine, equiaxed grain structure (average grain size: 102 ␮m) 2 min after inoculation. The present work illustrates that the production process for Al–Ti–B master alloys can be improved further by fine tuning the salt addition parameters.

[6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22]

A. Cibula, J. Inst. Met. 76 (1949) 321–346. B.S. Murty, S.A. Kori, M. Chakraborty, Int. Mater. Rev. 47 (2002) 3–29. D.G. McCartney, Int. Mater. Rev. 34 (1989) 247–260. K.V.S. Prasad, B.S. Murty, P. Pramanik, M. Chakraborty, Mater. Sci. Tech. 12 (1996) 766–770. C.D. Mayes, D.G. McCartney, G.J. Tatlock, Mater. Sci. Tech. 9 (1993) 97–103. A. Ohno, I. Motege, Solidification Technology in the Foundry and Casthouse, The Metals Society, London, 1983, pp. 171–175. A. Hardman, F.H. Hayes, Mater. Sci. Forum 217–222 (1996) 247–252. G.P. Jones, J. Pearson, Metall. Trans. 7B (1976) 223–234. M.S. Lee, B.S. Terry, Mater. Sci. Tech. 7 (1991) 609–612. Y. Miyasaka, Y. Masuda, US Patent No. 3,857,705, 1974. P. Davies, J.L.F. Kellie, D.P. Parton, J.V. Wood, US Patent No. 6,228,185, 2001. D.K. Young, W.C. Setzer, F.P. Koch, R.A. Rapp, M.J. Pryor, N. Jarrett, US Patent No. 5,415,708, 1995. P.S. Mohanty, J.E. Gruzleski, Acta Metall. Mater. 43 (1995) 2001–2012. A.L. Greer, P.S. Cooper, M.W. Meredith, W. Schneider, P. Schumacher, J.A. Spittle, A. Tronche, Adv. Eng. Mater. 5 (2003) 81–91. M. Easton, D. StJohn, Metall. Mater. Trans. 36A (2005) 1911–1920. N. Iqbal, N.H. van Dijk, S.E. Offerman, M.P. Moret, L. Katgerman, G.J. Kearley, Acta Mater. 53 (2005) 2875–2880. M.M. Guzowski, G.K. Sigworth, D.A. Sentner, Metall. Trans. 18A (1987) 603–619. L. Arnberg, L. Backerud, H. Klang, Met. Technol. 9 (1982) 7–13. I. Maxwell, A. Hellawell, Acta Metall. 23 (1975) 895–899. Y. Birol, J. Alloys Compd. 420 (2006) 71–76. B.S. Murty, S.A. Kori, K. Venkateswarlu, R.R. Bhat, M. Charaborty, J. Mater. Process. Technol. 89/90 (1999) 152–158. M.S. Lee, P. Grieveson, Scand. J. Metall. 32 (2002) 256–262.