Structural Homogeneity of Direct-Chill Cast Ingots of Aluminum Alloy EN AW-5083

Structural Homogeneity of Direct-Chill Cast Ingots of Aluminum Alloy EN AW-5083

Dolic, Natalija

Structural homogeneity of direct-chill (DC) cast ingots of aluminum alloy EN AW-5083 was investigated in terms of grain size and grain distribution using the Latin square experimental design. The ingot cross-sectional homogeneity, the grain sizes, and the mean grain number per unit area were determined at precise, statistically defined locations in the slice by means of a semiautomatic method for measuring mean lineal intercept lengths. Based on the analysis of the differences in the number and distribution of grains between the slices cut from the ingot front and those from its rear, a general assessment of the ingot structural homogeneity was made. Analysis of variance showed the highly significant differences, in grain number in specimens taken from the ingot front section to be related to slice height/ingot depth and in those taken from the ingot rear section to individual charges and slice height. The grand means of the mean number of grains per unit area for the ingot front and rear sections show relatively high values with respect to ingot size. The obtained correlation coefficient, which suggests a good agreement between the number and distribution of grains at the ingot front and those at its rear, is indicative of good structural homogeneity of the ingot in general.

DOI: 10.1007/s11663-006-9008-z

© The Minerals, Metals & Materials Society and ASM International 2007

OUR objective was to determine the structural homogeneity of ingots of aluminum alloy EN AW-5083, designated according to EN 573-3,’11 which were cast by the vertical direct-chill (DC) casting process on a reconstructed and revamped casting line,’ ’31 by measuring the size and the mean number of grains per unit area. The investigation was planned to follow the statistical design of the Latin square.’41 Sampling was done from slices of about 3 cm in depth, cut transversally from the front and rear sections of the cast ingots.

Investigation of structural homogeneity was carried out in six ingots cast by the semicontinuous vertical DC process. The ingots 1430×520×5100 mm in size were manufactured from six different charges of alloy EN AW-5083 (designations 3112, 3113, 3114, 3116, 3117, and 3120). Before casting, the melt was refined with an argon and chlorine mixture in an Alpur unit. For grain refinement, the Al-Ti5-B1 master alloy was used in an average amount of 1.9 kg/t melt; small bars of the alloy were added to the casting furnace, and a wire was introduced in a launder positioned in front of the Alpur unit.

Figure 1 is a schematic representation of the Latin square-based sampling design, which belongs to a group of orthogonal models that find large application when no interactions between variability sources-process factors are expected,[4] where i is the slice height/ingot depth and j is the slice width/ingot width. The specimens cut from the slice carry the following designation: charge number-specimen number, letter F or R. The letter F refers to specimens taken from the ingot front section and R refers to specimens taken from its rear.

To identify the grain boundaries, the specimens were subjected to electrolytic etching (anodization) with Barker’s reagent.[5,6] Viewed under polarized light, with the addition of a sensitive tint filter, the grains were well discernible and appeared as different shades of the appropriate colors.[7,8] The mean grain size was determined using a semiautomatic method for mean lineal intercept length measurement (the intercept procedure).[7,8,9]

Table I shows the chemical compositions of the examined charges determined by optical emission spectrometry. Specimens were taken during casting, from ingots of about 0.5 m in length.

Figures 2 and 3 are graphic representations of the relationship between the individual variables (charge, slice height/ingot depth, and slice width/ingot width) and the mean number of grains per unit area in EN AW-5083 specimens.

From the relationship between the individual variables and the mean number of grains per unit area, in the ingot front and rear sections, the following conclusions can be drawn.

(1) There is little variation in the mean number of grains per unit area per charge in respect to the grand mean as concerns the ingot front, whereas in the case of the ingot rear section, the charge appears to exert a significant effect because high deviations from the grand mean can be noted in either direction.

(2) The slice height-grain number relationship follows a distinct level-oriented pattern for both the ingot front and rear sections. The mean number of grains per unit area tends to increase with distance from the slice center (i = 1) toward the edge (i = 6), demonstrating a significant effect of the ingot cooling rate on grain distribution by slice height/ingot depth.

(3) The slice width-grain number relationship shows very similar patterns for the ingot front and rear sections. The tendency of the mean number of grains per unit area to grow higher alongside the ingot edges is taken to be the result of enhanced heat extraction from the ingot outer region and of consequential intensified undercooling setting off the solidification mechanism.

The variance analysis is analysis of variability between the means, expressed as the variance. The results obtained by the F-test for the mean number of grains per unit area for the ingot front and rear sections are shown in Tables III and IV.

From Table III, where the results of variance analysis are given for the ingot front, it is evident that the differences in the mean number of grains per unit area are highly significant in respect to slice height (P

From the results of variance analysis in Table IV pertaining to the ingot rear section, the differences in the number of grains per unit area in respect to charge and slice height appear to be highly significant (P

The fact that the mean number of grains per unit area range by slice width fails to show any significant variation is taken to be the result of efficacious inoculation and restricted grain growth. The latter is considered to be due to the alloying action of magnesium as well as of the other alloying elements taking part in the process of constitutional undercooling. However, data concerning the effects of the interaction between the solute elements and of the growth restricting factors are lacking.’111 On the other hand, the intensity of heat extraction from the ingot, i.e., its cooling rate, appears to be a predominant factor responsible for the change in the mean number of grains per unit area by slice height/ingot depth. The variability of the mean number of grains per unit area observed among individual charges can also be accounted for by slight differences in their chemical compositions.

The coefficient of correlation between the number and distribution of grains in the slices taken from the ingot front and rear sections was determined by Mest. Its value, r = 0.76, which is suggestive of a relatively high conformity of the number and distribution of grains between the two ingot sections, is a sign of a rather high structural homogeneity of the ingot in general, in other words, of the appropriately chosen refining and casting technologies and efficacious melt inoculation.

REFERENCES

1. prEN 573-3: 2003: Aluminium and Aluminium Alloys-Chemical Composition and Form of Wrought Products-Part 3: Chemical Composition, European Commitee for Standardization, 2003.

2. J. Prgin, D. Filipi, A. Markotic, and F. Unkic: Proc. 5th Int. Foundrymen Conf: Innovative Foundry Materials and Techologies, Opatija, Croatia, 2004, pp. 182-93.

3. D.G. Altenpohl: Aluminium: Technology, Applications and Enviroment, A Profile of a Modern Metal, Aluminium from Within, The Aluminum Association and TMS, Warrendale, PA, 1999, pp. 77-125.

4. F. Robert and P.E. Brewer: Design of Experiments for Process Improvement and Quality Assurance, Engineering & Management Press, Institute of Industrial Engineers, Norcross, GA, 1996, 11549.

5. E 407-99: Standard Practice for Microetching Metals and Alloys, American National Standard Institute/ASTM, Philadelphia, PA, 1999.

6. G F. Vander Voort: Metallography, Principles and Practice, 2nd ed., ASM INTERNATIONAL, Materials Park, OH, 1999, pp. 610-16.

7. N Dolic, K. Terzic, J. Prgin, A. Markotic, and F. Unkic: Proc. 6th Int. Foundrymen’s Conf: Inovative Foundry Materials and Technologies, Opatija, Croatia, 2005, pp. 160-74.

8. N Dolic, K. Terzic, J. Prgin, A. Markotic, and F. Unkic: Proc. Int. Conf: Materials, Tribology, Processing, MATRIB 2005, Vela Luka, Croatia, 2005, pp. 301-08.

9. E 1382-97: Standard Test Methods for Determining Average Grain Size Using Semiautomatic and Automatic Image Analysis, American National Standard Institute/ASTM, Philadelphia, PA, 1997.

10. E 112-96: Standard Test Methods for Determining Average Grain Size American National Standard Institute/ASTM, 2000.

11. T.E.’ Quested: Ph.D. Thesis, University of Cambrige, Cambnge, United Kingdom, 2004, pp. 5-53.

NATALIJA DOLIC, Research Assistant, ANTE MARKOTIC, Professor, and FARUK UNKIC, Associate Professor, are with the Faculty of Metallurgy, University of Zagreb, 44 103 Sisak, Croatia. Contact e-mail: ndolic@simet.hr

Manuscript submitted April 25, 2006.

Article published online May 12, 2007.

Copyright Minerals, Metals & Materials Society and ASM International Jun 2007

Provided by ProQuest Information and Learning Company. All rights Reserved