13th International Conference on Aluminum Alloys (ICAA13)
Edited by: Hasso Weiland, Anthony D. Rollett, William A. Cassada
TMS (The Minerals, Metals & Materials Society), 2012
ROLE OF SOLUTE AND TRANSITION METALS IN GRAIN
REFINEMENT OF ALUMINUM ALLOYS UNDER ULTRASONIC MELT
TREATMENT
L. Zhang1, 2, D.G. Eskin3, A. Miroux2 and L. Katgerman1
1
Delft University of Technology, Department of Materials Science and Engineering; Mekelweg
2; 2628CD Delft, The Netherlands
2
Materials innovation institute; Mekelweg 2; Delft 2628CD, The Netherlands
3
Brunel University, BCAST; Uxbridge; UB8 3PH, U.K.
Keywords: Grain refinement, Ultrasound, Aluminum alloy, Growth restriction factor
Abstract
It is well known that several factors affect the as-cast microstructure through influencing either
nucleation or growth during solidification. The effect of solute elements on grain refinement,
which can be quantified by the growth restriction parameter Q, is of great importance for
controlling the as-cast grain size in aluminum alloy. On the other hand, the ultrasonic melt
treatment (UST) is one of potent grain refining methods in aluminum alloy. Our previous results
showed that ultrasonic melt treatment results in a significant grain refining effect in an aluminum
alloy when transition metals, such as Ti and Zr, are present in the melt. It was suggested that
UST may increase the nucleating potency of Al3(Zr,Ti) intermetallics. In this paper, the growth
restriction effect caused by segregating elements is considered in order to further study the
efficiency and potency of grain refiner particles caused by ultrasonic treatment. The results show
that the main effect of transition metals Zr and Ti is related to the maximum number of particles
which are active nucleants under UST. With increasing Zr, the maximum number of potent
particles increases, which produces a better grain refining effect.
Introduction
Grain refinement in aluminum casting has been studied for many decades. A fine grain structure
is well known to benefit both the mechanical and technological properties of aluminum alloys [1,
2]. Several factors affect the as-cast microstructure through influencing either nucleation or
growth during solidification. The effect of solute elements on grain refinement, which can be
quantified by the growth restriction factor (GRF) Q, is of great importance for controlling the ascast grain size in aluminum alloy [3–6].
Generally, GRF represents the segregating power of all elements during solidification in the
alloy. These segregating elements accumulate at the solid/liquid interface, preventing its further
growth and, at the same time, lead to the formation of a constitutionally undercooled zone in
front of the interface, which might activate the potent nucleants present there [3–6]. GRF Q is
defined as ™ imiC0,i (ki–1), where for each i element, m is the liquidus gradient, C0 is the bulk
alloy composition of the alloy, and k is the partition coefficient (CS/CL, where CS and CL are the
compositions of the solid and liquid phases in the equilibrium). Through GRF Q, a simple
relation between grain size d and the alloy composition for a constant set of casting conditions is
suggested by Easton and StJohn [7–9] as
d = a + b/Q
(1)
The constant a is related to the maximum number of particles which are active nucleants and the
gradient b determines the nucleant potency of the particles.
Ultrasonic melt treatment is known to induce grain refinement in aluminum alloys. High
frequency and high amplitude oscillations result in cavitation of the melt and also promote
1389
intense mixing through agitation [10, 11]. The reported mechanisms are related either to
increasing the number of nucleation sites or to improving the nucleant potency of the particles
[10–15]. Our previous results show that when transition metals Zr and Ti are present in the melt,
UST becomes very efficient for grain refinement in aluminum alloys, especially when it is
performed in the temperature range of primary solidification of Al3Zr [12–15]. The ultrasonicaided grain refinement in the alloys containing Zr and Ti was related to the refinement of
primary Al3(Zr,Ti) intermetallics upon cavitation, significant increase of their number and
further involvement in solidification as nucleation sites for aluminum. The specific role of Ti in
Zr-contained aluminum alloy under UST has also been studied previously. Dissolution of Ti in
Al3Zr intermetallics was shown to change the lattice parameters of the intermetallics and delay
the nucleation of aluminium to lower temperatures. The larger resulting undercooling allows
small intermetallics, previously refined by UST, to become nucleation sites. The main effect of
Ti was therefore assigned to the improving the nucleation potency of refined Al3Zr intermetallics,
while UST causes the increase in the number of nucleating particles [14, 15].
In this paper, the specific role of Zr+Ti combination under ultrasonic melt treatment is studied at
different solute concentration in 7XXX-series aluminum alloys. By changing Zr and solute (Cu,
Mg and Zn) concentrations, six 7XXX-series alloys with different Q values were cast with and
without UST. The aim of this article is to discuss the effect of Zr and solute concentration on
grain refinement under UST.
Experimental procedure
In this research, six 7XXX-series alloys with two concentrations of Zr and three concentrations
of main alloying elements were prepared and cast with and without UST. Ti concentration was
constant in all studied alloys, around 0.06 wt%, in order to study independently the Zr effect.
Experimental alloys were prepared using 99.97 wt% pure Al, pure Zn, Al–6 wt% Ti, Al–6 wt%
Zr, Al–50 wt% Cu, and Al–25 wt% Mg master alloys.
The alloys firstly were molten in an electric furnace, and then treated in the clay–graphite
crucible (appr. 0.3 kg melt) at 700 °C by ultrasound for about 10 seconds. After UST, the melt
was poured into a preheated cupper mold. The cooling rates during solidification in the copper
mold were measured by a K-thermocouple as 2 r 0.1 K/s. Without ultrasonic treatment the
pouring melt temperature was 700 °C. The ultrasonic equipment consisted of a 5-kW generator,
5-kW water-cooled transducer, and a Nb sonotrode. Experiments were performed at 4.0 kW
generator power. The sonotrode was preheated before UST. The casting conditions and alloy
compositions are summarized in Table 1.
Table 1. The casting conditions and alloy compositions in this research
Composition, wt%
Additions,
Casting
Cooling rate,
Alloy
wt%
temperature, °C
K/s
Cu
Mg
Zn
0.13Zr, 0.06Ti
Low solute
1.6
1.3
4.3
content alloy
0.2Zr, 0.06Ti
0.13Zr, 0.06Ti
Medium solute
700
2
2.3
1.5
6.1
content alloy
0.2Zr, 0.06Ti
0.13Zr, 0.06Ti
High solute
3.5
1.8
8.3
content alloy
0.2Zr, 0.06Ti
The growth restriction factor Q for each alloy was calculated using the data [16, 17].
After casting, the samples were mechanically ground, polished, and electro-oxidized at 20 VDC
in a 3% HBF4 water solution. The grain structure was examined in an optical microscope under
1390
polarized light. Grain size was measured using the linear intercept method and the statistical
analysis of the results was performed. The chemical composition of all tested Al alloys was
measured using a spark spectrum analyzer (Spectro, Kleve, Germany).
Results and discussion
It is well known that in aluminum alloys, transition metals, such as Ti and Zr, are usually added
as graining and anti-recrystallization agents. In the case of UST, the role of Zr as a grain refining
elements was demonstrated before [10, 12–15]. In order to study the specific role of Zr under
UST, 0.2 wt% Zr and 0.13 wt% Zr with constant amount of Ti were added into three 7XXXseries Al alloys with different Q values, respectively.
Fig. 1 shows the effect of solute content on grain size of 7XXX-series Al alloys with 0.2 wt% Zr
and 0.06 wt% Ti processed by UST. It clearly shows that with 0.2 wt% Zr, UST results in good
grain refining for all tested alloys with different Q values. With increasing the solute content, the
grain size decreased. The quantitative information is given in Fig. 2.
Figure 1. Grain structure in 7XXX-series alloys with 0.2 wt % Zr and 0.06 wt% Ti: (a), (d) low
solute content alloy, without (a) and with (d) UST; (b), (e) medium solute content alloy, without (b)
and with (e) UST; (c), (f) high solute content alloy, without (c) and with (f) UST.
Figure 2. Effect of solute content on grain size of
7XXX-series Al alloys with 0.2 wt% Zr and 0.06
wt% Ti processed by UST.
Figure 3. Relationship between grain size and
1/Q in 7XXX-series Al alloys with 0.2 wt% Zr
and 0.06 wt% Ti.
1391
In order to analyze how UST affects the number of nucleation sites and nucleant potency of the
particles, we re-plotted the grain size from Fig. 2 with respect to Q values, as shown in Fig. 3.
The intercept a in Eq. (1) decreases when UST is used, which means that the maximum number
of particles that are active nucleants increases due to UST. This is in good agreement with our
previous results [12–15]. Meanwhile, the gradient b decreased slightly under UST as well, which
shows the nucleant potency of the intermetallics is also improved by UST.
In the alloys with 0.13 wt% Zr and 0.06 wt% Ti, the grain size only decreased in the alloys with
higher Q values and when UST was applied, as illustrated in Fig. 4–6. The ultrasonic grain
refining effect in these alloys is not as good as at the higher Zr concentration.
Figure 4. Grain structure in 7XXX-series alloy in addition to 0.13 wt % Zr and 0.06 wt% Ti: (a),
(d) low solute content alloy, without (a) and with (d) UST; (b), (e) medium solute content alloy,
without (b) and with (e) UST; (c), (f) high solute content alloy, without (c) and with (f) UST.
Figure 5. Effect of solute content on grain size of
7XXX-series Al alloys with 0.13 wt% Zr and 0.06
wt% Ti processed by UST.
Figure 6. Relationship between grain size and
1/Q in 7XXX-series Al alloys with 0.13 wt% Zr
and 0.06 wt% Ti.
By plotting the grain size with respect to the Q values (Fig. 6), we can see that the grain size
remains almost constant in non-treated alloys. UST, however, results in the increased amount of
nucleating particles, as demonstrated in Fig. 6 by the line with small intercept a. It might be
confusing in view of the interpretation of Eq. (1) [7, 8] that the alloys with 0.13 wt% Zr and cast
without UST seem to have more potent nucleating particles, as the slope b (nucleating
undercooling) is small. It is quite possible that in these alloys that are very close to the peritectic
1392
point in the Al–Zr system, the formation of primary equilibrium Al3Zr phase is not occurring
under given casting conditions and either few metastable L12 Al3Zr particles are formed (which
have very good crystallographic match with Al) or Zr is consumed by a supersaturated solid
solution. In the latter case, the gradient b represents the nucleant potency of the particles which
are naturally present in the melt. These particles might have good nucleating potency but are
scarce. UST has been reported that it facilitates the formation of equilibrium Al3Zr intermetallics
and dissolution of Ti in them [13–15]. It has been also shown that Ti being dissolved in Al3Zr
should actually increase the nucleating undercooling [15]. As a result the undercooling matches
the size of numerous refined particles and the efficient grain refinement occurs under UST.
It is worth to note that the Q values for the discussed alloys were only calculated by GRF of Cu,
Mg and Zn, not including Ti and Zr. On one hand, all Zr and Ti are assumed to be bound in
Al3(Zr, Ti) intermetallics during calculation. On the other hand, Ti is constant in all discussed
alloys above. The effect of Ti was considered to be equal in all studied 7XXX-series alloys.
Based on this, the role of Zr concentration and UST can be clearly shown in Fig. 7. As one can
see, under UST, the maximum number of potent intermetallics increases with increasing Zr
concentration, which produces a better grain refining effect, while the nucleant potency of the
particles remain the same. Through these results, we can conclude that three factors act in
achieving the optimum grain refining effect in Al alloys with Zr and Ti under UST. Firstly, the
amount of Zr should be sufficient to produce primary intermetallics and the UST should be
applied in the temperature range of their formation to assure their refinement. Secondly, Ti
should be added and be dissolved in Al3Zr so that the nucleating undercooling during
solidification will be adequate to the size of intermetallics. And thirdly, the solute elements have
additional role in this grain refinement by restricting the growth of aluminum grains. The role of
UST is mainly in increasing the number of nucleating particles.
Figure 7. Effect of Zr concentration and UST on the grain size at a constant Ti concentration.
Conclusions
The special role of Zr under UST has been studied in aluminum alloys with different growthrestriction factors. It was previously suggested that ultrasonic processing refines Al3(Zr,Ti)
primary intermetallics and increases the number of heterogeneous substrates and the nucleation
potency of these substrates. These effects result in significant grain refining in aluminum alloys.
This paper confirms this mechanism. The main effect of Zr addition is related to the number of
particles that become active nucleants under ultrasonic treatment. With increasing Zr
concentration, the number of particles increases which produces a better grain refining effect.
The solute elements act as an additional, growth restriction factor.
1393
Acknowledgments
This research was carried out under project number M11.5.11419 in the framework of the
Research Program of the Materials innovation institute M2i (www.m2i.nl). L. Zhang would like
to thank China Scholarship Council and Prof. J.S. Zhang for providing the scholarship for the
PhD study at TU Delft.
References
[1] B.S. Murty, S.A. Kori, and M. Chakraborty, “Grain refinement of aluminum and its alloys by
heterogeneous nucleation and alloying,” Int Mater Rev, 47 (2002), 3-29.
[2] E.R. Rooy, Metals handbook, vol.15: casting (Metals Park, OH: ASM international, 1989),
743-770.
[3] M.A. Easton, and D.H. StJohn, “The effect of alloy content on the grain refinement of
aluminium alloys,” TMS Light Metals, 2001, 927-933.
[4] A.L. Greer et al., “grain refinement of aluminium alloys by inoculation,” Adv Eng Mater, 5
(2003), 81-91.
[5] T. Chandrashekar et al., “Effect of growth restricting factor on grain refinement of aluminum
alloys,” Int J Adv Technol, 40 (2009), 234-241.
[6] T.E. Quested, and A.L. Greer, “Grain refinement of Al alloys: Mechanisms determining ascast grain size in directional solidification,” Acta Mater, 53 (2005), 4643-4653.
[7] D.H. StJohn et al., “New approach to analysis of grain refinement,” Int J Cast Met Res, 20
(2007), 131-135.
[8] M.A. Easton, and D.H. StJohn, “An analysis of the relationship between grain size, solute
content, and the potency and number density of nucleant particles,” Metall Mater Trans A, 36A
(2005), 1911-1920.
[9] M.A. Easton, and D.H. StJohn, “Improved prediction of grain size of aluminum alloys that
includes the effect of cooling rate,” Mater Sci Eng A, 486 (2008), 8-13.
[10] G.I. Eskin, Ultrasonic Treatment of Light Alloy Melts (Amsterdam: Gordon and Breach
science publisher, 1998)
[11] O.V. Abramov, Ultrasound in Liquid and Solid Metals (Boca Raton, FL: CRC Press, 1994)
[12] L. Zhang et al., “On the mechanism of the formation of primary intermetallics under
ultrasonic melt treatment in an Al-Zr-Ti alloy,” IOP Conf Ser: Mater Sci Eng, 27 (2012), 012002
[13] L. Zhang, D.G. Eskin, and L. Katgerman, “Influence of ultrasonic melt treatment on the
formation of primary intermetallics and related grain refinement in aluminum alloys,” J Mater
Sci, 46 (2011), 5252-5259.
[14] T.V. Atamanenko et al., “Criteria of Grain Refinement Induced by Ultrasonic Melt
Treatment of Aluminum Alloys Containing Zr and Ti,” Metall Mater Trans A, 41A (2010),
2056-2066.
[15] T.V. Atamanenko et al., “On the mechanism of grain refinement in Al–Zr–Ti alloys,” J
Alloys Comp, 509 (2011), 57-60.
[16] M. Johnsson, and L. Bäckerud, “The influence of composition on equiaxed crystal growth
mechanisms and grain size in Al alloys,” Z Metall, 87(3) (1996), 216-220.
[17] T.E. Quested, A.T. Dinsdale and A.L. Greer, “Thermodynamic modeling of growthrestriction effect in aluminum alloys,” Acta Mater, 53 (2005), 1323-1334.
1394