Вы находитесь на странице: 1из 7

Journal of Materials Processing Technology 162–163 (2005) 435–441

Grain refinement of LM6 Al–Si alloy sand castings to enhance


mechanical properties
Lim Ying Pio ∗ , Shamsuddin Sulaiman, Abdel Majid Hamouda,
Megat Mohamad Hamdan Megat Ahmad
Department of Mechanical and Manufacturing Engineering, Faculty of Engineering, Universiti Putra Malaysia,
43400 UPM Serdang, Selangor, Malaysia

Abstract

Al–Ti–B master alloys are well accepted in Al–Si alloys casting industry as grain-refiners to enhance the castings quality. Grain refinement
is achieved in aluminium alloys by inoculating the melt in the furnace with specific amount of selected grain-refiner. The grain-refiners come
in different chemical compositions and there is no guideline to add the optimal level of them into the melt for sound castings. This research
work investigates the effect of Al5Ti1B on the mechanical properties of LM6 Al–Si alloy sand casting. The typical LM6 aluminium alloy
contains 10–13 wt.% of silicon and thus inherently solidifies with coarse grain sizes. The mechanical properties ascertained are hardness and
ultimate tensile strength (UTS). The grain macrostructures of the castings are studied by optical and scanning electron microscopes. The
experimental work is performed on a sand casting of different modulus, which inherently induces different cooling rates to enable a simple
correlation between cooling rate and grain refinement level. The addition level of Al5Ti1B into the melt ranges from 0 wt.% to 1 wt.% with
the increment of 0.25 wt.%. The experimental results show that the mechanical properties of LM6 sand casting can be optimally improved by
grain refinement of 0.5 wt.% AL5Ti1B. Further increase of grain-refiner quantity does not provide any more significant improvement. The data
shows that solidification rate is directly proportional to the addition level of grain-refiners but inversely proportional to the casting modulus. In
the original sand casting of LM6, section of lower modulus with higher solidification rate has better mechanical properties. When the optimal
level of grain-refiner is added, more uniform mechanical properties are achieved throughout the casting irrespective of section modulus.
© 2005 Elsevier B.V. All rights reserved.

Keywords: Grain refinement; Sand casting; Hardness; Ultimate tensile strength; Solidification rate

1. Introduction in the casting, which otherwise will solidify with coarse


columnar grain structure. Fine equiaxed grain structure leads
Grain refinement is considered to be one of the most to several benefits, such as uniform distribution of second
important and popular melt treatment processes for phases and microporosity, improved feeding ability [5], high
aluminum–silicon alloys castings. The use of grain-refiners yield strength, high toughness, improved machinability and
to improve castings mechanical properties is widespread excellent deep drawability of the products [6].
in aluminium industry, and its associated benefits on final Several mechanisms take place in the formation of grains
products are well documented [1]. Grain refinement of in a casting during solidification. In general, there are two fac-
aluminium alloys provides a number of technical and tors that contribute to the formation of grains. First, there must
economic advantages, including reduced ingot cracking, be the presence of suitable substrates in sufficient amount to
better ingot homogeneity [2], susceptibility to hot cracking act as heterogeneous nucleation sites. Secondly, there has to
is reduced [3] and mechanical properties are improved be sufficient undercooling to facilitate the survival and growth
significantly [4]. Grain refinement improve the quality of the nuclei [7]. The undercooling can be achieved by either
of castings by reducing the size of primary ␣-Al grains rapid cooling to generate bulk undercooling and/or by parti-
tion of solute during solidification to generate constitutional
∗ Corresponding author. undercooling. Both of these criteria have to be fulfilled to
E-mail address: yingpio lim@yahoo.com (L.Y. Pio). obtain a small grain size in a casting [8].

0924-0136/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jmatprotec.2005.02.217
436 L.Y. Pio et al. / Journal of Materials Processing Technology 162–163 (2005) 435–441

As is well known, grain refinement by inoculation of melts


is based on the mechanism of heterogeneous nucleation [9].
The inoculation of aluminium alloys is usually achieved by
the addition of Al–Ti or Al–Ti–B master alloy into the melt of
base metal. Grain-refiners are produced commercially with
a wide range of Ti/B ratios. The Al–(3 or 5)%Ti–1%B and
Al–(3 or 5)%Ti–0.2%B grain-refiners of the Al–Ti–B sys-
tem have long dominated the grain refining market for sev-
eral decades. Several new types of grain-refiners had entered
the market in recent years, such as TiBAlloy and Strobloy
[10]. TiBAlloy contains 1.6%Ti, 1.4%B, 0%Sr and the boride Fig. 1. Mold system.
phase is majority in (Al,Ti)B2 . While Strobloy contains the
similar contents of titanium and boron, and additional 10%Sr.
The boride phases of Strobloy are TiB2 and SrB6 , and its shaped one with 20 mm diameter. The four ingates leads into
aluminide phase is SrAl4. The use of TiBAlloy for the hy- the cavity have similar cross-section of 19 mm × 7 mm, and
poeutectic alloys used in the foundry sector has been shown the runner has a cross-section of 15 mm × 20 mm. The feeder
to provide benefits in terms of minimizing porosity and lack was also designed in conical shape of top diameter 40 mm,
of fade [11]. Strobloy has been described as combining the bottom diameter 20 mm and height 50 mm. Initially the di-
benefits of grain refinement of TiBAlloy with the modifica- mensions were determined according to AFS standard guide-
tion effects of Sr [12]. Another new member of the grain- lines and specifications found in the literature but later it has
refiner family is Al–3%Ti–0.15%C. It has been used in a been found that the calculation results were not practical to
wide range of alloy systems and solidification process meth- construct a mold without blockage of feeding [15]. Therefore,
ods [13]. Compared to the traditional Al–Ti–B grain-refiners, the dimensions of the gating system as per the calculations are
a number of advantages are achieved by Al–3%Ti–0.15%C modified and replaced by the above-mentioned dimensions.
include melt cleanliness, interactions with degassing and fil- The overall design of the mold is shown in Fig. 1.
tration systems, finer grain structure and better surface ap- The pattern of the feeder, cavity and gating system are
pearance. The earliest systems in which Al–3%Ti–0.15%C made of wood [16]. The mold box or flask used to assemble
has found approval and acceptance to replace other grain- the cope and drag is also made of wood. Fig. 2 shows the
refiners have been in aerospace applications (7XXX series pattern and sand mold for the experimental analysis.
alloys) [14]. The tensile test samples were molded by another mold
In the present study, the grain-refiner employed is Al5Ti1B as shown in Fig. 3. Each mold will produce four pieces of
master alloy supplied by KBM AFFILIPS and the base metal test samples for similar chemical compositions of casting
is LM6. Previous casting with LM6 without grain-refiner is materials. The dimensions of the test samples are decided
used to benchmark the effect of Al5Ti1B grain refinement on according to ASTM E8-00.
the interested mechanical properties of harness and ultimate
tensile strength (UTS). The addition levels of grain-refiner
in terms of weight percentage are 0.25%, 0.5%, 0.75% and
1.0%, respectively. The casting process is CO2 sand casting
process.

2. Experimental procedures

The experimental procedures for making of molds, pour-


Fig. 2. Pattern and sand mold.
ing, testing and analysis of castings are explained in the fol-
lowing consecutive sections.

2.1. Mold design and construction

The casting was designed to have varying modulus (2.25,


4.08, 6.90 and 8.96) to incur different cooling rates during so-
lidification process. The thicknesses opted are 5 mm, 10 mm,
20 mm and 30 mm, respectively. The width and length of
each section are 80 mm and 100 mm. The sprue was designed
in conical shape with top diameter 28 mm, bottom diameter
18 mm and height 126 mm. The sprue well is a hemisphere Fig. 3. Tensile test sample mold.
L.Y. Pio et al. / Journal of Materials Processing Technology 162–163 (2005) 435–441 437

2.2. Mold materials and molding process Table 1


LM6 chemical compositions
Silica sand is the base material for making CO2 mold. The Chemical constituents wt.%
binder used is liquid sodium silicate solution [17]. The mold Al 84.95–87.95
is subjected to CO2 gassing for hardening before the pattern Cu 0.1
is stripped from the mold. This is called carbon dioxide sand Fe 0.6
Mg 0.1
molding. The sand is rammed manually by using a sand ram- Mn 0.5
mer over the sides of the pattern. After gassing with carbon Ni 0.1
dioxide for about 15 min, the pattern is detached from the Pb 0.1
sand. Cope and drag were assembled together by applying Si 10–13
glue at the parting line to prevent flashing of molten metal Sn 0.05
Ti 0.2
during pouring. Zn 0.1
Total four castings were produced with different addition levels of Al5Ti1B
2.3. Melting process and casting materials master alloy. The samples are designated as S000, S025, S050, S075 and
S100 according to 0%, 0.25%, 0.5%, 0.75% and 1% of Al5Ti1B added to
The melting of LM6 aluminium–silicon alloys was done the melt. The samples were polished and etched for microstructural analysis.
by an electrical resistance melting furnace. Temperature of
melt in the furnace is controlled and maintained at 750 ◦ C sile test machine and then subjected to tensile test. The data
before grain-refiner was subsequently added. The melt was are classified into the subsequent sections for discussion.
transferred to a ladle. From the ladle, the liquid alloy was
poured into the sand molds of cast and tensile test sample 3.1. Cooling curve and solidification rate
in sequence. LM6 is a corrosion resistant aluminium casting
alloy with average durability and strength, and also possess For the section of similar modulus, the cooling curves of
high impact strength and ductility. The chemical composition higher addition level of grain-refiner are noticed to be in-
of the melted alloy is analyzed by a spectrometer and the creasingly suppressed. That is, they solidify with faster cool-
various constituents present is tabulated in Table 1. ing rates when more grain-refiners are added into the melt.
For the typical section of modulus 2.25, this phenomenon is
shown in Fig. 4.
3. Experimental results and discussion The solidification rate is observed to be influenced by the
addition level of grain-refiner. When more grain-refiner is
K-type thermocouples and DataTaker were used to mea- added, the solidification rate will increase accordingly in all
sure and collect temperature data of the castings throughout sections of different modulus. The effect is plotted in Fig. 5
the mold filling and solidification processes. Test samples below.
were taken from each section of different modulus and sub- It is noticed that, the solidification rate drops at 0.75%
ject to hardness testing. The tensile test samples were grinded addition level of grain-refiner, and then raises up to maximum
to produce good surface finish for chuck gripping at the ten- at 1% grain refinement. This phenomenon occurs consistently

Fig. 4. Cooling curve of various addition level of grain-refiner for modulus 2.25.
438 L.Y. Pio et al. / Journal of Materials Processing Technology 162–163 (2005) 435–441

Fig. 5. Solidification rate vs. grain-refiner addition level by modulus.

Table 2 3.2. Hardness and ultimate tensile strength


Linear equation to correlate solidification rate and grain-refiner addition level
Modulus Least squares equations Fig. 7 shows the variation of ultimate tensile strength
2.25 SLR = 0.432 + 0.276GRL versus grain-refiner addition level. It is observed that the
4.08 SLR = 0.328 + 0.212GRL non-refined LM6 has an UTS of 76.73 MPa, after grain
6.90 SLR = 0.148 + 0.236GRL refinement of 0.5 wt.% Al5Ti1B, the UTS is improved by
8.96 SLR = 0.144 + 0.152GRL
39% to 106.68 MPa. Further increase of inoculation level
above 0.5 wt.% will not improve the mechanical properties
for all modulus. Linear equations computed by least squares infinitely. The results suggest that LM6 Al–Si alloy has the
approximation are determined for each modulus as listed in best response to Al5Ti1B grain refinement at 0.5 wt.% to
Table 2. The equations correlate solidification rate, K/s (SLR) enhance its ultimate tensile strength.
to grain-refiner level, wt.% (GRL) specifically for LM6 sand As indicated in Fig. 8, the hardness property also shows
castings as plotted in Fig. 6. the similar characteristics as ultimate tensile strength when

Fig. 6. Graph determined by least square method for solidification rates vs. grain-refiner addition levels.
L.Y. Pio et al. / Journal of Materials Processing Technology 162–163 (2005) 435–441 439

Fig. 7. Ultimate tensile strength vs. grain refinement level.

Fig. 8. Effect of grain-refiner on harness.

grain-refiner is added. The optimal wt.% of grain-refiner to be grain structure has become finer after the inoculating the LM6
added is 0.5 to achieve the highest hardness and no significant base metal with grain-refiner master alloy. It is generally well
improvement is observed above this level of addition. The accepted that the inoculation of master alloy into Al–Si melt
effect of grain refinement on LM6 castings is apparent, it will provide more heterogeneous nucleation sites for solid
has increased the hardness of the castings to be above 60 ␣-Al and result in finer grain size [20]. The microstructure
Rockwell HRD.
The literatures report that grain refinement of aluminium
alloys with various grain-refiners of Al3B, Al1Ti3B, Al3Ti
and Al5Ti1B also can effectively improve wear resistance
characteristics of Al and Al–Si alloys [18,19]. The wear rate
(mm3 /m) will decrease with the increase of grain-refiner ad-
dition level.

3.3. Grain structure

Figs. 9 and 10 shows the scanning electron microscope


(SEM) micrographs of original LM6 casting and that op-
timally refined with 0.5 wt.% Al5Ti1B. It is evident that the Fig. 9. Microstructure of non-refined LM6.
440 L.Y. Pio et al. / Journal of Materials Processing Technology 162–163 (2005) 435–441

Table 3
EDS spectrum analysis of LM6 castings
Element LM6 0.5 wt.% Al5Ti1B grain refinement

wt.% at.% wt.% at.%


O 9.79 15.56 3.34 5.52
Al 81.16 76.5 88.34 86.64
Si 8.46 7.66 8.33 7.84
Fe 0.59 0.27
Total 100 100

Fig. 10. Microstructure of 0.5 wt.% Al5Ti1B refined LM6.


It is observed that the grain-refined LM6 casting has higher
is fully equiaxed and no columnar grain is observed. Me- weight percentage of aluminium element by 7.18% compared
chanical testing highlights that when the grain size is refined, with the original LM6 casting. The silicon content remains
the mechanical properties of LM6 castings are consequently almost the same. Grain refinement with 0.5 wt.% Al5Ti1B
enhanced. will add 0.025 wt.% of titanium and 0.005 wt.% of boron into
The SEM possesses an energy dispersive spectroscopy the LM6 casting. These amounts are too low to be detected by
(EDS) facility. The EDS spectrum for original LM6 and the EDS system in use. According to literature, the titanium
0.5 wt.% Al5Ti1B grain-refined LM6 castings are shown in and boron generally exist in the forms of Al3 Ti and TiB2 ,
Figs. 11 and 12 below. Their respective element analysis is they could be difficult to be detected by EDS in relatively
shown in Table 3. small amount [21].

Fig. 11. EDS spectrum of original LM6 casting.

Fig. 12. EDS spectrum of 0.5 wt.% grain-refined LM6 casting.


L.Y. Pio et al. / Journal of Materials Processing Technology 162–163 (2005) 435–441 441

4. Conclusion als Processing, USA Publishers, Cambridge, MA, USA, 28–30 June
2000, 2001, pp. 181–188.
[4] D. Apelian, G.K. Sigworth, K.R. Whaler, Assessment of Grain
It is found that the inoculation of LM6 Al–Si alloy with
Refinement and Modification of Al–Si Foundry Alloys by Ther-
grain-refiner Al5Ti1B will improve the mechanical proper- mal Analysis, American Foundrymen’s Society Inc., 1984, pp.
ties of the castings. The hardness can be improved about 10% 99297–99307.
and the ultimate tensile strength also achieves a significant [5] D.G. McCartney, Grain refining of aluminium and its alloys using
improvement of 39%. The optimal addition level of grain- inoculants, Int. Mater. Rev. 34 (1989) 247–260.
[6] J.A. Spittle, J.M. Keeble, M.A. Meshhedani, The grain refinement
refiner is 0.5% to achieve the highest values of hardness and
of Al–Si foundry alloys, Light Met. (1997) 795–800.
ultimate tensile strength in the castings. Grain refinement [7] D.M. Stefanescu, Science and Engineering of Casting Solidification,
also increases the solidification rates of LM6 sand castings Kluwer Academic/Plenum Publishers, New York, 2002.
for all sections of different modulus. Faster solidification rate [8] M.N. Binney, D.H. StJohn, A.K. Dahle, J.A. Taylor, E.C. Burhop,
coupled with more heterogeneous nucleation sites will pro- P.S. Cooper, Grain refinement of secondary aluminium–silicon cast-
ing alloys, Light Metals (2003).
mote the formation of finer equiaxed grain sizes as indicated
[9] J. Campbell, Castings, Butterworth Heinemann, 1991.
in Fig. 10. The experimental data concludes that LM6 sand [10] P. Cooper, A. Hardman, E. Burhop, Characterization of a new gen-
castings can be effectively grain-refined with Al5Ti1B by eration of grain refiners for the foundry industry, Light Met. (2003).
introducing the master alloy in bar form into the melt just [11] T. Sagstad, N. Dahle, E. Bondhus, H. Koch, Grain refining of hy-
before pouring into the mold. The most optimally potent ad- poeutectic aluminium–silicon alloys with TiBloy, Molten Aluminium
Process. (AFS-1998) 97–116.
dition level is 0.5 wt.% to achieve the best mechanical prop-
[12] E. Bondhus, T. Sagstad, Strobloy: the new combined grain refiner
erties of hardness and ultimate tensile strength for LM6 sand and modifier for hypoeutectic Al–Si alloys, TMS Light Met. (2000)
castings. 845.
[13] A.M. Detomi, A.J. Messias, S. Majer, P.S. Cooper, The impact of
TiCAl and TiBAl grain refiners on casthouse processing, in: TMS
Acknowledgements 2001 Exhibition and Conference, New Orleans, LA, 11–15 February
2001.
[14] C.M. Small, et al., Microstructure and grain refining efficiency of
I would like to thank the KMP laboratory technicians for TiC particles in Al–Ti–C grain refining master alloys, in: Proceedings
their helps in performing sand casting and mechanical prop- of the ICAA6, 1998, p. 213.
erties testing. A sincere gratitude is also extended to Institute [15] R.W. Heine, C.R. Loper, P.C. Rosenthal, Principles of Metal Casting,
of Bioscience and Institute of Advanced Technology Univer- McGraw Hill, NY, 1978.
[16] A. Roosz, H.E. Exner, in: T.S. Piwonka, V. Voller, L. Katgerman
siti Putra Malaysia (ITMA) for using their facilities of optical
(Eds.), Modeling of Casting, Welding and Advanced Solidification
and scanning electron microscopes. Processes VI, TMS, Warrendale, PA, USA, 1993, p. 243.
[17] A.J. Clegg, Precision Casting Processes, Pergamon Press, UK, 1991.
[18] A.K. Prasada Rao, K. Das, B.S. Murty, M. Chakraborty, Effect of
References grain refinement on wear properties of Al and Al–7Si alloy, Wear
(2003).
[1] T. Sagstad, N. Dahle, et al., Grain refining of hypoeutectic [19] M. Harun, I.A. Talib, A.R. Daud, Effect of element additions on
aluminium–silicon alloys with TiBloy, in: Proceedings of the Fifth wear property of eutectic aluminium–silicon alloys, Wear 194 (1996)
International AFS Conference on Molten Aluminium Processing, 54–59.
American Foundrymen’s Society, Orlando, 1998, pp. 100–116. [20] C. Limmaneevichitr, W. Eidhed, Fading mechanism of grain refine-
[2] B. Chalmers, The structure of ingots, J. Aust. Inst. Met. 8 (1963) ment of aluminium–silicon alloy with Al–Ti–B grain refiners, Mater.
255–263. Sci. Eng. A349 (2003) 197–206.
[3] S.A. Metz, M.C. Flemings, A fundamental study of hot tearing, in: [21] S.A. Kori, B.S. Murty, M. Chakraborty, Development of an efficient
The Merton C. Flemings Symposium on Solidification and Materi- grain refiner for Al–7Si alloy, Mater. Sci. Eng. A280 (2000) 58–61.

Вам также может понравиться