Materials Letters 214 (2018) 276–279

Contents lists available at ScienceDirect

Materials Letters
journal homepage: www.elsevier.com/locate/mlblue

Improved ductility and toughness of an Al-Cu casting alloy by changing

the geometrical morphology of dendritic grains
Qing-Long Zhao, Qing-Quan Zhang, Wei Zhang, Feng Qiu, Qi-Chuan Jiang ⇑
Key Laboratory of Automobile Materials, Ministry of Education, and Department of Materials Science and Engineering, Jilin University, No. 5988 Renmin Street, Changchun 130025,
PR China

a r t i c l e i n f o a b s t r a c t

Article history: Different dendritic grain morphology of an Al-Cu alloy was obtained by adjusting the casting parameters.
Received 21 October 2017 Compared to the equiaxed dendritic grains, the tortuous dendritic grains led to increased tensile elonga-
Received in revised form 6 December 2017 tion (from 10.4% to 16.8%) and work of fracture (from 11.2 J/mm2 to 17.2 J/mm2) by more than 50%,
Accepted 8 December 2017
respectively. Meanwhile, the high tensile strength (540 MPa) and grain size (80 lm) were unchanged.
Available online 9 December 2017
The intergranular fracture crack zigzagged along tortuous grain boundaries, reducing the stress concen-
tration at the crack tip and increasing the propagation resistance and path, leading to increased tough-
Keywords:
ness and elongation. The zigzag morphology of grain boundaries was attributed to tortuous dendritic
Metals and alloys
Grain boundaries
grains.
Deformation and fracture Ó 2017 Elsevier B.V. All rights reserved.
Microstructure

1. Introduction retaining composition, grain size and mechanical strength

Traditional high-strength aluminum casting alloys usually exhi-
bit low ductility and toughness, which limits their application [1].
The fracture of aluminum casting alloys is usually initiated at
coarse second-phase particles and grain boundaries, because of
particle cracking or rupturing of eutectic matrix on grain bound-
aries [2,3]. Previous research mainly focused on the effect of grain
size, second dendrite arm spacing and coarse particles on fracture,
however, many efforts have been made to improve the ductility
and toughness of aluminum casting alloy by grain refinement
and heat treatment [4,5]. Meanwhile, limited research reported
that tortuous grain boundary morphology influenced ductility-
dip cracking during hot deformation and stress corrosion cracking
in Ni alloys [6,7]. The change of dendritic grain morphology might
also affect the ductility/toughness of cast aluminum alloys.
Cast Al-Cu alloys exhibited high strength, but its low ductility
and toughness limit its application. In our previously studies, we
found that a dendritic grain structure with medium grain size in
Al-Cu exhibited higher elongation than the more refined grain
structure, but its mechanism was unclear [8–10]. The present work
reports a significant improvement in ductility and toughness of an
Al-Cu casting alloy due to tortuous morphology of dendritic grains,
⇑ Corresponding author.
E-mail address: jqc@jlu.edu.cn (Q.-C. Jiang).
unchanged. The microstructure and fracture behavior were exper-
imentally investigated. The present work suggests a new approach
of improving the ductility and toughness of cast alloys.
2. Experimental
The chemical composition of the Al-Cu alloy (in wt.%) was
mainly Al-5Cu-0.45Mn (containing a small amount of other ele-
ments). 0.5 wt.% TiC nano-particles were added in the melt as grain
refiner, and the casting procedure has been described in Ref. [9].
Two different molds were used to produce different grain mor-
phology. The geometry of molds was illustrated in the Supplemen-
tary Material. T6 treatment was performed (12 h at 808 K and 10 h
at 438 K). The tensile specimens were with a gauge length of 10
mm and a cross section of 5 mm  2.5 mm. Tensile tests were per-
formed by a servo-hydraulic material testing system (MTS 810) at
room temperature with a strain rate of 3.0  10 4 s 1. Three-point
bend test was performed using an Instron materials testing
machine (Instron 5689), and the sample size was 16  4  2 mm.
A U-notch of 0.25 mm wide and 2 mm long was made in the mid-
dle of the longitudinal section. Every sample was polished before
three-point bend testing. Fracture toughness parameters were cal-
culated according to ref. [11]. The microstructure was observed by
scanning electron microscopy (SEM, TESCAN VEGA 3) equipped
with electron backscatter diffraction detector (EBSD). The EBSD
data were analyzed by HKL Channel 5 software.

https://doi.org/10.1016/j.matlet.2017.12.037
0167-577X/Ó 2017 Elsevier B.V. All rights reserved.
 Q.-L. Zhao et al. / Materials Letters 214 (2018) 276–279 277

3. Results and discussion
Fig. 1(a) shows the mechanical strength of two samples was
similar, as shown in Table 1. The total elongation of sample S2
was 16.8%, 1.6 times higher than that of sample S1 (10.4%). Fig.1
(b) shows that the peak load of sample S2 was similar to that of
sample S1, but the load dropped more slowly in sample S2 than
in S1 after the peak load. The linear-elastic plane-strain fracture
toughness of sample S1 and S2 was similar (Table 1), but the
elastic-plastic plane-strain fracture toughness and work of fracture
of sample S2 were 14.3% and 56% higher than those of sample S1,
respectively.
Fig. 2(a–b) indicates that both of samples S1 and S2 showed
intergranular fracture. The macro-crack in three-point bent sample
S1 was relatively straight and sharp, while the primary crack in
sample S2 was diffuse, and considerable deformation zone was dis-
tributed widely around the primary crack, as shown in Fig. 2(c–d).
The area ahead of the crack tip was examined by EBSD (Fig. 2e–f).
The strain distribution was analyzed by the local misorientation
method of grain average misorientation (GAM) [12,13]. Strain
was severely concentrated in the grains ahead of the crack tip in
sample S1 (Fig. 2e). On the contrast, the grains ahead of the crack
tip in sample S2 were less strained, and the strain concentrated
at grain boundaries (Fig. 2f). The grains were classified into three
groups according to the internal misorientation (h), as shown in
Fig. 2(e–f). Heavily deformed grains (h greater than 3°) was
observed at the crack tip in S1, corresponding to the strain concen-
trated grains, while there were almost no heavily deformed grains

Fig. 1. (a) Tensile stress-strain curves; (b) load – indenter displacement curves of sample S1 and S2 by three-point bend testing.

Table 1
Typical mechanical properties including yield strength (r0.2), tensile strength (rb), total elongation (e), linear-elastic plane-strain fracture toughness (KI); elastic-plastic plane-
strain fracture toughness (KJ); work of fracture (cwof).

Sample r0.2 rb e KI KJ cwof

(MPa) (MPa) (%) (MPa
m1/2) (MPa
m1/2) (J/mm2)
S1 310 ± 3 540 ± 3 10.4 ± 0.8 39.2 ± 0.3 71.3 ± 0.5 11.2 ± 0.5
S2 313 ± 5 545 ± 3 16.7 ± 1.2 39.1 ± 0.2 81.5 ± 0.5 17.5 ± 0.3

Fig. 2. Fracture surface morphology of sample S1 (a) and S2 (b) after tension testing; macro-cracks of sample S1 (c) and S2 (d) as ceased at the displacement = 0.4 mm in Fig.
2b; the tip of main crack and the corresponding EBSD analysis of sample S1 (e) and S2 (f).
278 Q.-L. Zhao et al. / Materials Letters 214 (2018) 276–279

Fig. 3. The metallography of crack propagation in (a) sample S1 and (b) S2 after three-point bending test; grain boundary 2D straightness distribution of sample S1 and S2 (c);
the grain structure in (d) sample S1 and (e) S2 after T6, and showing the measure method of grain boundary 2D straightness; the dendritic grain morphology of (f) sample S1
and (g) S2; (h) cooling curves.

in sample S2. The fraction of undeformed grains (h < 1°) in sample the higher cooling rate of sample S1. The lower cooling rate of S2
S2 was 42.6%, much higher than that in S1 (6.2%), indicating less might retard the start of poisoned growth, leading to the smaller
strain concentration in sample S2 as well. width of dendrite arms in S2 (12 lm) than that in S1 (18 lm).
As shown in Fig. 3(a–b), both of the two samples showed mainly Finally, TiC nano-particles might play an important role as well.
intergranular fracture. The primary crack of sample S1 was rela- TiC nano-particles promote heterogeneous nucleation of a-Al [9],
tively straight along the loading direction, while that in sample resulting in reduced neighbor interaction distance. The solutal
S2 was zigzag and deviated more from the loading direction, which interaction is enhanced, and then the effect of the different cooling
led to a lowered resolved shear stress for crack propagation. Thus conditions on dendrite growth becomes evident.
the crack propagation speed would be lowered in sample S2
(Fig. 1b). Moreover, the zigzagged crack increased the actual crack
4. Conclusions
length of sample S2, which consumed more fracture energy.
Straightness k was measured from the metallography to quantita-
The tortuous dendritic grain morphology with a large fraction of
tively compare the geometrical morphology of dendritic gains in
zigzag grain boundaries were formed in an Al-5Cu casting alloy by
the two samples, as shown in Fig. 3(d–e). The details of measure-
adjusting casting process, which led to significantly increased ten-
ment were described in the Supplementary Material. The average
sile elongation by 50%, retaining the grain size and mechanical
straightness of sample S1 was 0.321, smaller than that of sample
strength almost unchanged, compared to equiaxed dendritic grains
S2 was 0.433. Fig. 3(c) demonstrates that the straightness in sam-
with smooth grain boundaries. The intergranular crack propagated
ple S1 was generally below 0.5, while 30% of k value in sample S2
along the zigzag grain boundaries, leading to significantly
was larger than 0.5, suggesting that a large proportion of grain
increased fracture toughness and work of fracture. The different
boundaries in sample S2 showed tortuous morphology. Thus, the
dendrite morphology obtained by using different material and
intergranular crack propagated along the tortuous grain bound-
geometry of mold, might be attributed to dendrite growth affected
aries, leading to increased work of fracture and toughness.
by cooling rate and solutal interaction with neighboring grains.
The grain boundary morphology is related to the dendrite struc-
ture formed during different solidification process [14]. Coarse pri-
Acknowledgments
mary dendritic arms but no evident secondary dendrite arms are
observed in sample S1 (Fig. 3f–g). In sample S2, primary dendrites
This work is supported by the National Natural Science Foundation
are slender, and a considerable amount of secondary dendrite arms
of China (NNSFC, No. 51571101 and No. 51601066), and the
retain. Fig. 3(h) shows cooling curves during solidification. Two
Science and Technology Development Program of Jilin Province,
curves start to evidently separate after recalescence. The difference
China (Grant No. 20160520116JH and 20170101215JC).
in cooling rate below 903 K is due to the material and geometry of
mold. The higher thermal effusivity of the mold for S1 leads to a
higher cooling rate, as indicated by our simplified calculation in Appendix A. Supplementary data
the Supplementary Material. The higher cooling rate might induce
higher undercooling, which affects primary dendrite formation. Supplementary data associated with this article can be found, in
Primary dendrite growth can be divided into free growth and poi- the online version, at https://doi.org/10.1016/j.matlet.2017.12.037.
soned growth regimes [15]. Poisoned growth manifested itself as
decelerating dendrite growth, when the effect of solutal interaction References
with neighboring grains is significant. It is suggested that higher
undercooling promotes poisoned dendrite growth, which leads to [1] S.G. Pantelakis, N.D. Alexopoulos, Mater. Des. 29 (2008) 80–91.
reduced neighbor-interaction distance and increased dendrite tip [2] A. Pineau, A.A. Benzerga, T. Pardoen, Acta Mater. 107 (2016) 424–483.
[3] M.F. Ashby, C. Gandhi, D.M. Taplin, Acta Mater. 27 (1979).
radius [15]. Dendrite tip radius affects the width of dendrite arms, [4] Y. Zhang, S. Ji, Z. Fan, J. Alloys Compd. 710 (2017) 166–171.
and hence the dendrite morphology in S1 might be attributed to [5] M. Zhu, Z. Jian, G. Yang, Y. Zhou, Mater. Des. 36 (2012) 243–249.
 Q.-L. Zhao et al. / Materials Letters 214 (2018) 276–279 279

[6] A.J. Ramirez, J.W. Sowards, J.C. Lippold, J. Mater. Process. Technol. 179 (2006)
212–218.
[7] Q. Peng, J. Hou, Y. Takeda, T. Shoji, Corros. Sci. 67 (2013) 91–99.
[8] D.S. Zhou, J. Tang, F. Qiu, J.G. Wang, Q.C. Jiang, Mater. Charact. 94 (2014) 80–85.
[9] D. Zhou, F. Qiu, Q. Jiang, Mater. Sci. Eng. A. 622 (2015) 189–193.
[10] W.S. Tian, D.S. Zhou, F. Qiu, Q.C. Jiang, Mater. Sci. Eng. A. 658 (2016) 409–414.
[11] Y. Wang, P. Shen, R.F. Guo, Z.J. Hu, Q.C. Jiang, Ceram. Int. 43 (2017) 3831–3838.
[12] S.I. Wright, M.M. Nowell, D.P. Field, Microsc. Microanal. 17 (2011) 316–329.
[13] T. Benjamin Britton, A.J. Wilkinson, Acta Mater. 60 (2012) 5773–5782.
[14] W. Kurz, D.J. Fisher, Fundamentals of solidification, forth ed., Trans Tech
Publication Lte, Switzerland, 1988, pp. 147–180.
[15] A. Bogno, H. Nguyen-Thi, G. Reinhart, B. Billia, J. Baruchel, Acta Mater. 61
(2013).