Owing to the presence of a large amount of fine (Fe,Mn)3AlC carbides within austenite () matrix, the tensile property of the
Fe-30%Mn-8.5Al%-2.0%C (in mass%) alloy in the as-quenched condition was clearly superior to that of the as-quenched FeMnAlC (C 5 1:3%) alloys investigated by previous workers. After being aged at 823 K for 3 h, the present alloy could possess high yield strength up to 1262 MPa with an excellent 32.5% elongation. With almost equivalent ductility, the yield strength obtained was about 16% higher than that of the FeMnAlC (C 5 1:3%) alloys after solution heat-treatment or controlled-rolling followed by an optimal aging at 823 K. Additionally, due to the pre-existing fine (Fe,Mn)3AlC carbides within the matrix in the as-quenched alloy, the aging time required for attaining the optimal combination of strength
and ductility was much less than that of the FeMnAlC (C 5 1:3%) alloys aged at 823 K. When the present alloy was aged at 823 K for a time period longer than 4 h, both the strength and ductility were drastically dropped due to the occurrence of o= (o: carbon-deficient austenite)
lamellar structure on the = grain boundaries. [doi:10.2320/matertrans.M2010013]
(Received January 15, 2010; Accepted March 19, 2010; Published April 28, 2010)
Keywords: iron-manganese-aluminum-carbon alloy, spinodal decomposition, tensile test, lamellar structure, ductility
1. Introduction
A lot of effort has been made to develop austenitic FeMnAlC alloys as high-strength, high-ductility alloy steels. In their studies, it is seen that when an alloy with a chemical composition in the range of Fe-(2634)%Mn-(711)%Al-(0:541:3)%C was solution heat-treated and then quenched, the microstructure of the alloy was single-phase austenite ().1–17) When the as-quenched alloy was aged at
temper-atures ranging from 773 to 873 K for moderate times, fine (Fe,Mn)3AlC carbides started to precipitate coherently within
the matrix, and no precipitates were found on the = grain boundaries. The resulting microstructure could possess
a remarkable combination of strength and ductility.9–14)
However, when the aging time was increased within the temperature range, a ! (ferrite) þ reaction, a ! þ -Mn reaction, a ! þ -Mn reaction, a ! þ
þ -Mn reaction, or a ! oþ reaction occurred on
the = grain boundaries.13–17,22) The phase is also an
(Fe,Mn)3AlC carbide, which was formed on the = grain
boundaries as a coarse particle. For convenience, 0 carbide
and carbide are used to represent the (Fe,Mn)3AlC carbide
formed coherently within the matrix and heterogeneously
on the = grain boundaries.11,13–17) The formation of the
coarse carbides and -Mn precipitates on the grain boundaries resulted in the embrittlement of the alloys.12–15,22)
In addition to the extensive studies of the austenitic FeMnAlC alloys with C 5 1:3%, the microstructural developments in the FeMnAlC alloys with C > 1:3%
have also been given in several literatures.18–20) In these
studies, it is seen that the as-quenched microstructure of Fe-(2930)%Mn-(7:79)%Al-(1:52:5)%C alloys was
phase containing fine 0 carbides.19,20)The fine 0 carbides
were formed within the matrix by spinodal decomposition
during quenching.20) This is quite different from that
observed in the austenitic FeMnAlC alloys with C 5 1:3%,
in which the fine 0 carbides could only be observed in the
aged alloys. When the as-quenched alloy was aged between
823 and 1073 K, the fine 0 carbides grew and a
heteroge-neous reaction of þ 0!
oþ occurred on the = grain
boundaries. With increasing aging time, the heterogeneous reaction became predominant. Consequently, the stable microstructure of the alloy was found to be a mixture of (oþ) lamellar structure.19,20)In contrast to the studies of
the microstructures, information concerning the mechanical properties of the austenitic FeMnAlC alloys with C = 1:3% is very deficient. We are aware of only one article,19)in which
the tensile properties of the Fe-26 at%Mn-15 at%Al-8 at%C (Fe-30%Mn-8.5%Al-2.0%C, in mass%) alloy were exam-ined. However, all of their examinations were only focused on the alloy aged at 1073 K. No information was provided for the alloy aged at lower temperatures. Therefore, the purpose of this work is an attempt to investigate the relationship between the microstructures and the tensile properties of the Fe-30%Mn-8.5%Al-2.0%C alloy in the as-quenched condition and aged at 823 K for various times.
2. Experimental Procedure
The Fe-30%Mn-8.5%Al-2.0%C (in mass%) alloy was prepared in a vacuum induction furnace by using 99.7% iron, 99.9% manganese, 99.9% aluminum and pure carbon powder. The melt was cast into a 20 30 100 mm steel mold. After being homogenized at 1523 K for 12 h under a protective argon atmosphere, the ingot was hot-rolled to a final thickness of 6 mm. The plate was subsequently solution heat-treated at 1473 K for 2 h and then quenched into room-temperature water rapidly. Aging processes were
*1Graduate Student, National Chiao Tung University
*2Corresponding author, E-mail: tfl[email protected], dir.mse93g@nctu.
performed at 823 K for various times in a vacuum furnace and then quenched into water. Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were used to examine the microstructures and the tensile fracture surface as well as free surface. TEM specimens were prepared by using a double-jet electropolisher with an electrolyte of 60% acetic acid and 40% ethanol. Tensile tests were carried out at room-temperature with an Instron
tensile testing machine at a strain rate of 5 104s1.
Tensile test specimens were plates having 50.0 mm gauge length, 12.5 mm width and 5.0 mm thickness. The yield strength was measured at 0.2% offset strain, and the percent elongation was the total elongation measured after fracture.
3. Results and Discussion
Transmission electron microscopy examination indicated that in the as-quenched condition, the microstructure of the
alloy was phase containing fine 0 carbides. A typical
microstructure is shown in Fig. 1(a). Figure 1(b), a selected-area diffraction pattern taken from a mixed region covering
the matrix and fine 0 carbides, indicates that the fine 0
carbides have an L01
2 structure.16,17,20,21) In Fig. 1(b), it is
also seen that satellites lying along h100i reciprocal lattice directions about the (200) and (220) reflections could be observed. The existence of the satellites demonstrates that the
fine 0 carbides were formed during quenching by spinodal
decomposition.20) This is similar to that observed by the
present workers in the as-quenched
Fe-30%Mn-9%Al-2.0%C alloy.20) Figure 2(a) is a bright-field (BF) electron
micrograph of the as-quenched alloy aged at 823 K for 3 h,
revealing that the fine 0carbides grew within the matrix,
and no precipitates could be observed on the grain bounda-ries. However, when the alloy was aged at 823 K for 4 h, some coarse precipitates started to appear on the = grain boundaries. An example is shown in Fig. 2(b). Electron diffraction examinations indicated that the coarse precipitates on the grain boundaries were carbides. With continued aging at 823 K, the coarse carbides grew into the adjacent
grains through a þ 0!
oþ reaction, as illustrated
in Fig. 2(c). Figure 2(d), a selected-area diffraction pattern (SADP) taken from an area covering the carbide marked as ‘‘K’’ in Fig. 2(c) and its surrounding ophase, indicates that
the orientation relationship between the carbide and o
phase is ½001k ½001oand ð100Þk ð100Þo. With
increas-ing agincreas-ing time at 823 K, the þ 0!
oþ reaction would
proceed toward the whole austenite grains. Consequently, the stable microstructure of the present alloy at 823 K was
a mixture of (oþ) phases. A typical microstructure is
shown in Fig. 3. It is clear in Fig. 3(b) that the mixture of (ophase + carbide) has a lamellar structure.
Figure 4 shows the tensile properties of the present alloy in the as-quenched condition and aged at 823 K for various times. In Fig. 4, it is seen that the as-quenched alloy has ultimate tensile strength (UTS) 1105 MPa, yield strength (YS) 883 MPa, and an excellent 54.5% elongation. After being aged at 823 K for 3 h, the alloy can possess the highest UTS (1395 MPa) and YS (1262 MPa) with a good elongation
of 32.5%, which may be attributed to the growth of the 0
carbides within the matrix and no precipitates on the grain boundaries. When the alloy was aged at 823 K for 4 h, both of the strength and elongation were slightly decreased. This result was due to the over-coarsening of the 0carbides within the matrix and the presence of a small amount of
Fig. 1 Transmission electron micrographs of the as-quenched alloy: (a) BF, (b) an SADP taken from the mixed region of austenite matrix and fine 0carbides. The foil normal is [001] (hkl: phase; hkl: 0carbide), and (c) ð100Þ
carbides on the grain boundaries. However, when the alloy was aged at 823 K for 5 h, the elongation was drastically dropped from about 30.5% to 19.6%. In order to clarify why the elongation was drastically dropped, fracture and free surface analyses were undertaken by using SEM. Figure 5(a), a fractograph of the alloy aged at 823 K for 4 h, reveals that the alloy had a ductile dimple fracture surface, and some dimples contain one or more carbides (as indicated by arrows). Figure 5(b) is a SEM micrograph taken from the free surface contiguous to the fracture surface, indicating that slip bands were generated over the specimen, and some isolated microvoids were formed along the grain boundaries (as indicated by arrows). It is clearly seen in Fig. 5 that the structure had a high resistance to crack propagation and exhibited self-stabilization under deformation. However, the fracture surface of the alloy aged at 823 K for 6 h revealed
Fig. 2 (a) through (c), BF electron micrographs of the alloy aged at 823 K for (a) 3 h, (b) 4 h, and (c) 5 h, respectively. (d) an SADP taken from the carbide marked as ‘‘K’’ in (c) and its surrounding ophase. The foil normal is [001] (hkl: ophase; hkl: carbide).
Fig. 3 Scanning electron micrographs of the alloy aged at 823 K for (a) 6 h, and (b) 7 h, respectively.
Fig. 4 Tensile properties of the alloy in the as-quenched condition and aged at 823 K for various times.
largely cleavage facets as well as intergranular fracture and a few dimples, as shown in Fig. 6(a). Figure 6(b) demonstrates that microcracks (as indicated by arrows) were observed only at coarse carbides within the o= lamellar structure and no
cracks could be observed in the þ 0regions. Therefore, it
is reasonable to believe that the existence of the o= lamellar
structure would be mainly responsible for the crack initiation, which led to the drastic drop of the ductility.
On the basis of the preceding results, some discussion is appropriate. According to the previous studies in the Fe-(2634)%Mn-(711)%Al-(0:541:3)%C-(01:75)%(Nb+
V+Mo+W) alloys,1–17,21–25)it is seen that the as-quenched
microstructure was single phase or phase with (Nb,V)C carbides. Depending on the chemical composition, the alloys in the as-quenched condition show various UTS ranging from 814 to 993 MPa, YS ranging from 423 to 552 MPa
and elongation from 72 to 50%.4–9,21)By the optimal aging
treatments between 773 and 873 K for moderate times, a high
density of fine 0 carbides started to precipitate coherently
within the matrix and no precipitates were formed on the grain boundaries. The resulting microstructure could lead to the optimal combination of the mechanical strength and ductility. With an elongation better than about 30%, the values of 9531259 MPa for UTS and 6651094 MPa for YS could be attained.9–14,21–25)Obviously, owing to contain a
large amount of fine 0 carbides within the matrix, the
tensile property of the present alloy in the as-quenched condition is not only superior to that of the as-quenched
Fe-(2634)%Mn-(711)%Al-(0:541:3)%C-(01:75)%(Nb+ V+Mo+W) alloys but comparable to that of the aged alloys. Furthermore, when the present alloy was aged at 823 K for 3 h, the UTS and YS could reach up to 1395 and 1262 MPa, respectively, with a good elongation of 32.5%. Compared to the previous studies, it is found that with almost equivalent ductility, the present alloy possesses yield strength about 16% higher than the Fe-(2634)%Mn-(711)%Al-(0:54 1:3)%C-(01:75)%(Nb+V+Mo+W) alloys after the solu-tion heat-treatment or controlled-rolling followed by the
optimal aging.9–14,21–25) The reason is probably that due to
higher carbon content in the present alloy, a greater amount
of 0 carbides could be formed within the matrix during
aging. Additionally, in the previous studies of the austenitic FeMnAlC (C 5 1:3%) alloys aged at 823 K, it was generally concluded that the aging time required for attaining the optimal combination of strength and ductility was about
1516 h.10–14,21–25) Whereas, the aging time of the present
alloy aged at 823 K was only about 3 h, which was attributed to the pre-existing fine 0carbides within the matrix in the
as-quenched condition.
Finally, it is worthwhile to point out that, in the previous study of the Fe-26 at%Mn-15 at%Al-8 at%C alloy (the chemical composition of the alloy is similar to that of the present alloy),19)Kimura et al. reported that when the alloy was solution heat-treated at 1373 K for 1 h and then furnace cooled to room temperature, the alloy exhibited almost zero ductility due to a lot of coarse carbides on the = grain
Fig. 5 Scanning electron micrographs of the alloy (aged at 823 K for 4 h) after tensile fractured. (a) fracture surface, and (b) free surface, respectively.
Fig. 6 Scanning electron micrographs of the alloy (aged at 823 K for 6 h) after tensile fractured. (a) fracture surface, and (b) free surface, respectively.
to deduce that the þ structure is thought to be more
favorable for both strength and ductility than the oþ
lamellar structure.
4. Conclusions
The relationship between the microstructures and the tensile properties of the Fe-30%Mn-8.5%Al-2%C alloy was investigated. The obtained results are as follows:
(1) The as-quenched microstructure of the present alloy
was phase containing fine 0 carbides. The fine 0
carbides were formed within the matrix by spinodal decomposition during quenching. The tensile property of the as-quenched alloy was far superior to that of the as-quenched FeMnAlC (C 5 1:3%) alloys.
(2) After being aged at 823 K for 3 h, the present alloy could possess a high yield strength up to 1262 MPa with an excellent 32.5% elongation. With almost equivalent ductility, the yield strength obtained was about 16% higher than that of the FeMnAlC (C 5 1:3%) alloys after solution heat-treatment or controlled-rolling fol-lowed by an optimal aging at 823 K.
(3) Due to the pre-existing fine (Fe,Mn)3AlC carbides
within the matrix in the as-quenched alloy, the aging time required for attaining the optimal combination of strength and ductility was much less than that of the FeMnAlC (C 5 1:3%) alloys.
(4) When the alloy was aged at 823 K for a time period longer than 4 h, the o= lamellar structure was formed
on the = grain boundaries. The o= lamellar
structure was mainly responsible for the crack initia-tion, which led to the drastic drop of the ductility.
3) T. F. Liu, J. S. Chou and C. C. Wu: Metall. Trans. A 21 (1990) 1891– 1899.
4) S. C. Tjong and S. M. Zhu: Mater. Trans. JIM 38 (1997) 112–118. 5) S. C. Chang, Y. H. Hsiau and M. T. Jahn: J. Mater. Sci. 24 (1989) 1117–
1120.
6) K. S. Chan, L. H. Chen and T. S. Liu: Mater. Trans. JIM 38 (1997) 420–426.
7) J. D. Yoo, S. W. Hwang and K. T. Park: Mater. Sci. Eng. A 508 (2009) 234–240.
8) H. J. Lai and C. M. Wan: J. Mater. Sci. 24 (1989) 2449–2453. 9) J. E. Krzanowski: Metall. Trans. A 19 (1988) 1873–1876. 10) K. Sato, K. Tagawa and Y. Inoue: Scr. Metall. 22 (1988) 899–902. 11) K. Sato, K. Tagawa and Y. Inoue: Mater. Sci. Eng. A 111 (1989) 45–50. 12) I. Kalashnikov, O. Acselrad, A. Shalkevich and L. C. Pereira: J. Mater.
Eng. Perform. 9 (2000) 597–602.
13) W. K. Choo, J. H. Kim and J. C. Yoon: Acta Mater. 45 (1997) 4877– 4885.
14) K. Sato, K. Tagawa and Y. Inoue: Metall. Trans. A 21 (1990) 5–11. 15) S. C. Tjong and C. S. Wu: Mater. Sci. Eng. 80 (1986) 203–211. 16) C. N. Hwang, C. Y. Chao and T. F. Liu: Scr. Metall. 28 (1993) 263–
268.
17) C. Y. Chao, C. N. Hwang and T. F. Liu: Scr. Metall. 28 (1993) 109– 114.
18) Y. Kimura, K. Hayashi, K. Handa and Y. Mishima: Mater. Sci. Eng. A 329(2002) 680–685.
19) Y. Kimura, K. Handa, K. Hayashi and Y. Mishima: Intermetallics 12 (2004) 607–617.
20) C. S. Wang, C. N. Hwang, C. G. Chao and T. F. Liu: Scr. Metall. 57 (2007) 809–812.
21) I. S. Kalashnikov, O. Aksel’rad and M. S. Khadyev: Metal. Sci. Heat. Treat. 48 (2006) 219–225.
22) G. S. Krivonogov, M. F. Alekseyenko and G. G. Solov’yeva: Fiz. Metal. Metalloved. 39 (1975) 775–781.
23) I. S. Kalashnikov, B. S. Ermakov, O. Aksel’rad and L. K. Pereira: Metal. Sci. Heat. Treat. 43 (2001) 493–496.
24) I. S. Kalashnikov, O. Acselrad, A. Shalkevich, L. D. Chumakova and L. C. Pereira: J. Mater. Process. Tech. 136 (2003) 72–79.
