- Email: [email protected]
Antiferromagnetic Mn50Fe50 wire with large magnetostriction
ARTICLE IN PRESS Journal of Magnetism and Magnetic Materials 321 (2009) 3778–3781
Contents lists available at ScienceDirect
Journal of Magnetism and Magnetic Materials journal homepage: www.elsevier.com/locate/jmmm
Antiferromagnetic Mn50Fe50 wire with large magnetostriction Aina He, Tianyu Ma, Jingjing Zhang, Wei Luo, Mi Yan State Key Laboratory of Silicon Materials, Department of Materials Science and Engineering, Zhejiang University, Hangzhou 310027, China
a r t i c l e in f o
a b s t r a c t
Article history: Received 22 April 2009 Received in revised form 12 July 2009 Available online 23 July 2009
This work presents a study on the relation between the fiber texture and the magnetostrictive performance in an antiferromagnetic Mn50Fe50 alloy wire, which was prepared through the combining process of hot rolling and cold drawing. The face-centered cubic (fcc) crystal structure can be retained during the plastic deformation process. Mixed fiber textures consisting of both /11 0S and /1 0 0S components were formed along the drawing direction (DD) in the wire. A large magnetostriction of 750 ppm was obtained along DD under 1.2 T, which can be ascribed to the single g phase and the formation of preferred crystal orientation. & 2009 Elsevier B.V. All rights reserved.
PACS: 75.80.+q 76.50.+g 75.50.Ee Keywords: Magnetostriction Mn–Fe wire Fiber texture Plastic deformation
1. Introduction Magnetostrictive wires, such as Ni, Fe–Ni and Fe–Ga alloy wires are of great interest for transducer and sensor applications [1,2]. The early magnetostrictive wires, such as Ni and Fe–Ni alloy wires have been stepped into applications due to their good mechanical properties, but they can only possess a relatively low magnetostriction of about 20 ppm [3,4]. It has been reported recently that a Fe85Ga15 alloy wire developed by combining plastic deformation processes can exhibit a room temperature magnetostriction of 66 ppm under a relatively low magnetic field of 80 Oe [2]. However, the high cost of Ga element should be a limitation for further applications. Although the terfenol alloys have also attracted much attention due to their giant magnetostriction over 2000 ppm, it is difficult to fabricate wire materials because of their brittleness [5,6]. In 2006, Peng and Zhang have reported that the magnetostriction of an antiferromagnetic gMn42Fe58 polycrystalline sample can reach 169 ppm under 1 T [7]. This value is comparable to that of ferromagnetic Fe–Ga alloys [8,9], although it requires a much higher magnetic field. Moreover, Mn–Fe alloys possess a high mechanical strength and good ductility, in addition to the low material cost [10,11]. It is of interest to develop g-Mn–Fe wires as potential candidates in magnetostrictive applications in transducers and sensors. Due to the good ductility, a textured g-Mn–Fe alloy has been prepared through cold rolling, exhibiting improved magnetostriction
Corresponding author.
E-mail address: [email protected] (M. Yan). 0304-8853/$ - see front matter & 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jmmm.2009.07.038
performance [12]. In this work, a Mn50Fe50 magnetostrictive wire was prepared through a combining plastic deformation process. The formation of the crystallographic texture along the wire axis and its relation to the magnetostrictive performance were also discussed.
2. Experimental Polycrystalline Mn50Fe50 alloy sample was fabricated by vacuum induction melting under an argon atmosphere using high purity iron and high purity electrolytic manganese. The ingot was remelted for three times to ensure homogeneity. Sample with the size of 8 10 100 mm3 was cut from the ingots. The combined processing of hot rolling and cold drawing was schematically shown in Fig. 1. The ingot was homogenized for 24 h at 900 1C, and then was rolled down to a thickness of 5 mm at 850 1C. The specimen was cut to 5 5 180 mm3. After thermal annealing at 700 1C for 4 h, cold drawing was undertaken to obtain a wire with a diameter of 3.6 mm. Phase identification and texture determination of the as-cast and the deformed Mn50Fe50 specimens were conducted by a D/max 2550 pc X-ray diffractometer with Cu Ka radiation, respectively. The wire specimens with a length of 1.5 mm were cut from the F 3.6 mm wire and put together into a ring with a diameter of 10 mm. The cross section of the wire specimens was mechanically ground and polished for pole figure measurement. The (2 0 0) and (2 2 0) pole figures were used to describe the fiber texture. The specimens were ground and subsequently etched in a
ARTICLE IN PRESS A. He et al. / Journal of Magnetism and Magnetic Materials 321 (2009) 3778–3781
reagent with 5 g ferric chloride, 2 ml hydrochloric acid and 100 ml alcohol. Microstructures of the etched samples were observed under an optical microscopy (LEICA MPS60). Magnetostriction measurements were performed on specimens with the size of 3 10 2 mm3 by standard strain gauge method. The external magnetic field was up to 1.2 T, which was applied along DD.
Fig. 1. Schematic diagrams of the combined processes of hot rolling and cold drawing.
Fig. 2. XRD patterns of Mn50Fe50 alloys: (a) the as-cast sample, (b) the as-rolled sheet, (c) the annealed sheet and (d) the as-drawn wire, respectively. (b) and (c) were taken from the sample surface parallel to the rolling plane; (d) was taken from the sample surface which lies parallel to the wire surface.
3779
3. Results Fig. 2 shows the XRD patterns of Mn50Fe50 alloys for the as-cast sample, the as-rolled sheet, the annealed sheet, and the as-drawn wire, respectively. It can be seen that all the samples exhibit single austenite (g) phase with a fcc structure, indicating that no stress-induced-structural-transformation occurred during the deformation processes of hot rolling and cold drawing. Fig. 3 shows optical micrographs of the longitudinal section of the as-cast and the as-drawn Mn50Fe50 samples. The as-cast Mn50Fe50 alloy had a regular equiaxed polycrystalline structure with an average grain size of about 65 mm and a random distribution of crystallographic orientations of the g phase, as shown in Fig. 3a. After hot rolling and cold drawing, a significantly deformed microstructure consisting of elongated grains was found and most of the elongated grains were arranged along DD, as shown in Fig. 3b. It can also be seen that the plastic deformation was inhomogenous with an average elongated grain size of about 150 mm long and 30 mm wide. Fig. 4 shows X-ray diffraction pole figures measured on the as-drawn Mn50Fe50 alloy wires. Textures measured using X-ray diffraction measurements are global textures. The pole figure presents, in a stereographic projection, using two dimension color contour maps, the distribution of a special crystal direction with respect to the sample reference axis, e.g., DD. The uniform distribution of poles represents random crystal orientation, while dense clusters of poles indicate strong textures. The {h k l} /u v wS designates the plane {h k l} which lies parallel to the wire surface and the direction /u v wS is parallel to DD. The fiber texture of the as-drawn Mn50Fe50 alloy wire, as shown in Fig. 4, from (2 0 0) and (2 2 0) pole figures, can be described as {0 11} /1 0 0S and {0 11} /11 0S, which were indicated by different symbols of the filled circles and delta-stars, respectively. Additionally, there was a texture component with a {0 0 1} /11 0S, which indicated by the filled squares. Therefore, /1 0 0S and /11 0S duplex fiber texture paralleled to DD was formed after cold drawing. Fig. 5 shows the magnetostriction as a function of magnetic field for the as-cast and the as-drawn samples at room temperature. The external magnetic field was applied along DD. It can be seen that the magnetostriction values of the samples increased steadily as the applied field increased, which were not saturated even under an applied field of 1.2 T. Magnetostriction value under 1.2 T for the as-cast Mn50Fe50 sample was just 303 ppm. After hot rolling and cold drawing, which induced a fiber texture along DD, magnetostriction under 1.2 T of the as-drawn sample reached 750 ppm, which was twice as high as that of the as-cast one.
Fig. 3. Optical micrographs of the longitudinal section of the (a) as-cast and (b) as-drawn Mn50Fe50 samples.
ARTICLE IN PRESS 3780
A. He et al. / Journal of Magnetism and Magnetic Materials 321 (2009) 3778–3781
Fig. 4. (2 0 0) and (2 2 0) pole figures measured on the as-drawn Mn50Fe50 alloy wire. DD denotes the drawing direction. Intensity contours in these pole figures are indicated by colors. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 5. Magnetostriction of the as-cast and as-drawn Mn50Fe50 samples measured along DD.
4. Discussion Mn–Fe alloys can exhibit as single g phase terminal solid solution alloys with a fcc structure for Mn contents in the range of 40–75 at% at room temperature, according to the Mn–Fe phase diagram [13]. In our previous work, it has been demonstrated that when Mn contents are below 40 at%, Mn–Fe alloys consisting of g and e phases possess much poor magnetostriction performance than the alloys with single g phase when Mn contents are above 40 at% [14]. It is also revealed that the magnetostrictive performance of Mn42Fe58 alloy deteriorates after annealing for 24 h at 1100 1C due to the separation of single g phase into a mixture of g, e and a phases [15]. These results suggest that single g phase in Mn–Fe alloys is essential to obtain large magnetostriction. In this work, the Mn50Fe50 alloy still remains a single g phase after hot rolling and cold drawing, ensuring the large magnetostriction. Plastic deformation occurs in metals by slip or by twinning on specific favorable atomic planes and crystallographic directions. In a polycrystalline sample, even though the initial orientation of grains is random, after sufficient deformation most of the grains
are realigned into a preferred orientation, hence developing a specific texture. From Fig. 4, it can be seen that the as-drawn Mn50Fe50 alloy wire shows a mixture of /1 0 0S and /11 0S fiber components. Combining with the microstructure observation (Fig. 3b), we put forward a primary explanation as follows. For the Mn50Fe50 alloy wire, a great amount of deformed grains are produced during rolling and drawing, and are arranged along DD. The nature of fiber textures in fcc metals mainly depends on the stacking fault energy (SFE) [16]. Jiang et al. have revealed that the fiber texture of the Fe–27Mn–6Si–5Cr alloy wire showed /1 0 0S and /11 0S duplex components [17]. The SFE of g phase in the Mn–Fe alloys with Mn atom ratio a (Mn) ¼ 20–50% could be calculated by embedded-atom method (EAM) as follows: SFE (mJ/m2) ¼ 23.3+0.269a (Mn)% [18]. According to this equation, the SFE in the g-Mn50Fe50 alloy is higher than that of g phase in the Fe–27Mn–6Si–5Cr alloy. Therefore, the /1 0 0S and /11 0S fiber textures parallel to DD is formed in the g-Mn50Fe50 alloy wire, which is the characteristic of fcc metals with a high SFE [19]. From Fig. 5, it can be seen that the magnetostriction of the drawn g-Mn50Fe50 alloy wire achieves 750 ppm under 1.2 T. The value is eleventh times as large as that of Fe–Ga wires [2], although it requires the use of two orders of magnitude at higher magnetic field. While the magnetostriction mechanism in antiferromagnetic g-Mn–Fe alloys remains unclear, owing to still lack in direct observation of antiferromagnetic domain structures in these alloys. However, the magnetostriction in an antiferromagnetic NiO crystal under an applied field, in principle, can be due to (i) the movement of twin boundaries, (ii) the alignment of antiferromagnetic domains, and (iii) the rotation of the spins [20]. For antiferromagnetic g-Mn50Fe50 alloy samples, there is no martensitic structure observed in the present work. That is to say, the absence of twin boundaries leads to a conclusion that the magnetostriction of the Mn50Fe50 alloy may be due to the movement of antiferromagnetism and/or the rotation of spins. Moreover, Peng and Zhang have predicted that the magnetostriction of Mn–Fe alloys could be affected by grain orientations and magnetic domain configurations [7]. Kennedy and Hicks have also demonstrated that the magnetic structure of the antiferromagnetic g-Mn–Fe alloys was a noncolinear spin structure with spin directions along /111S [21]. The magnetostriction of g-Mn–Fe alloys may be due to rotation of spins within a set of {111} planes. The mixture of /1 0 0S and /11 0S fiber components parallel to DD formed during deformation might result in the movement of domains and rotation of spins that becomes much easier during magnetizing
ARTICLE IN PRESS A. He et al. / Journal of Magnetism and Magnetic Materials 321 (2009) 3778–3781
process, which should be responsible for the magnetostrictive performance improvement. To further improve the magnetostriction of Mn–Fe alloys, single crystal growth might be a potential choice. Compared to the polycrystalline materials, improved magnetostriction has previously been achieved in single crystals, i.e. Terfenol-D and Ni–Mn–Ga alloys [5,22–24]. Saturation magnetostriction in polycrystalline Terfenol-D is just about 1000 ppm, but reaches 2375 ppm under a compressive pre-stress of 24 MPa and a magnetic field of 2.0 T in a single crystal prepared by Czochralski method. Meanwhile, single crystals of magnetic shape memory alloys Ni–Mn–Ga also possess much larger magnetostrain than the polycrystalline samples. For example, a giant magnetostrain of 6.3% can be achieved in a single crystal Ni50Mn27.5Ga22.5 with a five-layer (5M) structure, which is about one order of magnitude higher than the polycrystalline sample. Although it still lack in explanation of the magnetostrictive mechanism of antiferromagnetic Mn–Fe alloys, higher magnetostriction could be expected in single crystals. It is worth performing study on single crystal growth of Mn–Fe alloy in further work.
5. Conclusions An antiferromagnetic Mn50Fe50 alloy wire with large magnetostriction was successfully prepared by combining plastic deformation process through hot rolling and cold drawing. The retention of single g phase during deformation is essential to obtain large magnetostriction in the Mn50Fe50 alloy wire. A mixture of /1 0 0S and /11 0S duplex fiber texture along the drawing direction is formed in the as-drawn wire. Room temperature magnetostriction of up to 750 ppm under 1.2 T is achieved for the as-drawn Mn50Fe50 wire, which is twice as high as that of the as-cast state. Such large magnetostriction should be ascribed to the preferred crystal orientation along DD formed during plastic deformation.
3781
Acknowledgements This work was supported by the Natural Science Foundation of China (NSFC-50701039 and 50531010), Visiting Scholar Foundation of State Key Laboratory of Silicon Materials (SKL2009-6), and the Innovative Research Team in University (IRT-0651). References [1] K. Mohri, F.B. Humphrey, K. Kawashima, K. Kimura, M. Mizutani, IEEE Trans. Magn. 26 (1990) 1789. [2] J.H. Liu, F. Yi, C.B. Jiang, J. Alloys Compd. 481 (2009) 57. [3] C. Luna, V. Raposo, G. Rauscher, M. Va zquez, IEEE Trans. Magn. 38 (2002) 3108. [4] V. Raposoa, C. Lunaa, M. Va zquez, J. Magn. Magn. Mater. 249 (2002) 22. [5] A.E. Clark, Ferromagnetic Materials, Vol. 1, North-Holland, Amsterdam, 1980 P.531. [6] T.Y. Ma, M. Yan, J.J. Zhang, W. Luo, C.B. Jiang, H.B. Xu, Appl. Phys. Lett. 90 (2007) 102502. [7] W.Y. Peng, J.H. Zhang, Appl. Phys. Lett. 89 (2006) 262501. [8] S. Guruswamy, N. Srisukhumbowarnchai, A.E. Clark, J.B. Restorff, M. WunFogle, Scr. Mater. 43 (2000) 239. [9] J.H. Li, X.X. Gao, J. Zhu, C.X. He, J.W. Qiao, M.C. Zhao, J. Alloys Compd. 476 (2009) 529. [10] O. Grassela, L. Kruger, G. Frommeyer, L.W. Meyer, Int. J. Plast. 16 (2000) 1391. [11] A. Druker, C. Sobrero, H.G. Brokmeier, J. Malar\curr {r´}a, R. Bolmaro, Mater. Sci. Eng. A 481 (2008) 578. [12] T.Y. Ma, J.J. Zhang, A.N. He, M. Yan, Scr. Mater. 61 (2009) 427. [13] H.P. Kneijnsberg, C.A. Verbraak, M.J. Ten Bouwhuijs, Acta Metall. 33 (1985) 1759. [14] Y.W. Xu, T.Y. Ma, J.J. Zhang, M. Yan, Acta Metall. Sin. 44 (2008) 1235. [15] J.J. Zhang, T.Y. Ma, A.N. He, M. Yan, J. Alloys Compd. doi: 10.1016/j.jallcom. 2009.06.004. [16] A.T. English, G.Y. Chin, Acta Metall. 13 (1965) 1013. [17] B.H. Jiang, X. Qi, Y.L. Ren, C.Q. Wang, Mater. Trans. JIM 41 (2000) 663. [18] Y.H. Rong, Q.P. Meng, G. He, T.Y. Hsu, J. Shanghai Jiaotong Univ. 37 (2003) 171. [19] C.S. Lee, B.J. Duggan, Metall. Trans. A 22A (1991) 2637. [20] L. Alberts, E.W. Lee, Proc. Phys. Soc. London 78 (1961) 728. [21] K.J. Kennedy, T.J. Hicks, J. Phys. F: Met. Phys. 17 (1987) 1599. [22] G.H. Wu, X.G. Zhao, J.H. Wang, J.Y. Li, K.C. Jia, W.S. Zhan, Appl. Phys. Lett. 67 (1995) 2005. [23] R.C. O’Handley, J. Appl. Phys. 83 (1998) 3263. [24] C.B. Jiang, J.M. Wang, H.B. Xu, Appl. Phys. Lett. 86 (2005) 252508.
Contents lists available at ScienceDirect
Journal of Magnetism and Magnetic Materials journal homepage: www.elsevier.com/locate/jmmm
Antiferromagnetic Mn50Fe50 wire with large magnetostriction Aina He, Tianyu Ma, Jingjing Zhang, Wei Luo, Mi Yan State Key Laboratory of Silicon Materials, Department of Materials Science and Engineering, Zhejiang University, Hangzhou 310027, China
a r t i c l e in f o
a b s t r a c t
Article history: Received 22 April 2009 Received in revised form 12 July 2009 Available online 23 July 2009
This work presents a study on the relation between the fiber texture and the magnetostrictive performance in an antiferromagnetic Mn50Fe50 alloy wire, which was prepared through the combining process of hot rolling and cold drawing. The face-centered cubic (fcc) crystal structure can be retained during the plastic deformation process. Mixed fiber textures consisting of both /11 0S and /1 0 0S components were formed along the drawing direction (DD) in the wire. A large magnetostriction of 750 ppm was obtained along DD under 1.2 T, which can be ascribed to the single g phase and the formation of preferred crystal orientation. & 2009 Elsevier B.V. All rights reserved.
PACS: 75.80.+q 76.50.+g 75.50.Ee Keywords: Magnetostriction Mn–Fe wire Fiber texture Plastic deformation
1. Introduction Magnetostrictive wires, such as Ni, Fe–Ni and Fe–Ga alloy wires are of great interest for transducer and sensor applications [1,2]. The early magnetostrictive wires, such as Ni and Fe–Ni alloy wires have been stepped into applications due to their good mechanical properties, but they can only possess a relatively low magnetostriction of about 20 ppm [3,4]. It has been reported recently that a Fe85Ga15 alloy wire developed by combining plastic deformation processes can exhibit a room temperature magnetostriction of 66 ppm under a relatively low magnetic field of 80 Oe [2]. However, the high cost of Ga element should be a limitation for further applications. Although the terfenol alloys have also attracted much attention due to their giant magnetostriction over 2000 ppm, it is difficult to fabricate wire materials because of their brittleness [5,6]. In 2006, Peng and Zhang have reported that the magnetostriction of an antiferromagnetic gMn42Fe58 polycrystalline sample can reach 169 ppm under 1 T [7]. This value is comparable to that of ferromagnetic Fe–Ga alloys [8,9], although it requires a much higher magnetic field. Moreover, Mn–Fe alloys possess a high mechanical strength and good ductility, in addition to the low material cost [10,11]. It is of interest to develop g-Mn–Fe wires as potential candidates in magnetostrictive applications in transducers and sensors. Due to the good ductility, a textured g-Mn–Fe alloy has been prepared through cold rolling, exhibiting improved magnetostriction
Corresponding author.
E-mail address: [email protected] (M. Yan). 0304-8853/$ - see front matter & 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jmmm.2009.07.038
performance [12]. In this work, a Mn50Fe50 magnetostrictive wire was prepared through a combining plastic deformation process. The formation of the crystallographic texture along the wire axis and its relation to the magnetostrictive performance were also discussed.
2. Experimental Polycrystalline Mn50Fe50 alloy sample was fabricated by vacuum induction melting under an argon atmosphere using high purity iron and high purity electrolytic manganese. The ingot was remelted for three times to ensure homogeneity. Sample with the size of 8 10 100 mm3 was cut from the ingots. The combined processing of hot rolling and cold drawing was schematically shown in Fig. 1. The ingot was homogenized for 24 h at 900 1C, and then was rolled down to a thickness of 5 mm at 850 1C. The specimen was cut to 5 5 180 mm3. After thermal annealing at 700 1C for 4 h, cold drawing was undertaken to obtain a wire with a diameter of 3.6 mm. Phase identification and texture determination of the as-cast and the deformed Mn50Fe50 specimens were conducted by a D/max 2550 pc X-ray diffractometer with Cu Ka radiation, respectively. The wire specimens with a length of 1.5 mm were cut from the F 3.6 mm wire and put together into a ring with a diameter of 10 mm. The cross section of the wire specimens was mechanically ground and polished for pole figure measurement. The (2 0 0) and (2 2 0) pole figures were used to describe the fiber texture. The specimens were ground and subsequently etched in a
ARTICLE IN PRESS A. He et al. / Journal of Magnetism and Magnetic Materials 321 (2009) 3778–3781
reagent with 5 g ferric chloride, 2 ml hydrochloric acid and 100 ml alcohol. Microstructures of the etched samples were observed under an optical microscopy (LEICA MPS60). Magnetostriction measurements were performed on specimens with the size of 3 10 2 mm3 by standard strain gauge method. The external magnetic field was up to 1.2 T, which was applied along DD.
Fig. 1. Schematic diagrams of the combined processes of hot rolling and cold drawing.
Fig. 2. XRD patterns of Mn50Fe50 alloys: (a) the as-cast sample, (b) the as-rolled sheet, (c) the annealed sheet and (d) the as-drawn wire, respectively. (b) and (c) were taken from the sample surface parallel to the rolling plane; (d) was taken from the sample surface which lies parallel to the wire surface.
3779
3. Results Fig. 2 shows the XRD patterns of Mn50Fe50 alloys for the as-cast sample, the as-rolled sheet, the annealed sheet, and the as-drawn wire, respectively. It can be seen that all the samples exhibit single austenite (g) phase with a fcc structure, indicating that no stress-induced-structural-transformation occurred during the deformation processes of hot rolling and cold drawing. Fig. 3 shows optical micrographs of the longitudinal section of the as-cast and the as-drawn Mn50Fe50 samples. The as-cast Mn50Fe50 alloy had a regular equiaxed polycrystalline structure with an average grain size of about 65 mm and a random distribution of crystallographic orientations of the g phase, as shown in Fig. 3a. After hot rolling and cold drawing, a significantly deformed microstructure consisting of elongated grains was found and most of the elongated grains were arranged along DD, as shown in Fig. 3b. It can also be seen that the plastic deformation was inhomogenous with an average elongated grain size of about 150 mm long and 30 mm wide. Fig. 4 shows X-ray diffraction pole figures measured on the as-drawn Mn50Fe50 alloy wires. Textures measured using X-ray diffraction measurements are global textures. The pole figure presents, in a stereographic projection, using two dimension color contour maps, the distribution of a special crystal direction with respect to the sample reference axis, e.g., DD. The uniform distribution of poles represents random crystal orientation, while dense clusters of poles indicate strong textures. The {h k l} /u v wS designates the plane {h k l} which lies parallel to the wire surface and the direction /u v wS is parallel to DD. The fiber texture of the as-drawn Mn50Fe50 alloy wire, as shown in Fig. 4, from (2 0 0) and (2 2 0) pole figures, can be described as {0 11} /1 0 0S and {0 11} /11 0S, which were indicated by different symbols of the filled circles and delta-stars, respectively. Additionally, there was a texture component with a {0 0 1} /11 0S, which indicated by the filled squares. Therefore, /1 0 0S and /11 0S duplex fiber texture paralleled to DD was formed after cold drawing. Fig. 5 shows the magnetostriction as a function of magnetic field for the as-cast and the as-drawn samples at room temperature. The external magnetic field was applied along DD. It can be seen that the magnetostriction values of the samples increased steadily as the applied field increased, which were not saturated even under an applied field of 1.2 T. Magnetostriction value under 1.2 T for the as-cast Mn50Fe50 sample was just 303 ppm. After hot rolling and cold drawing, which induced a fiber texture along DD, magnetostriction under 1.2 T of the as-drawn sample reached 750 ppm, which was twice as high as that of the as-cast one.
Fig. 3. Optical micrographs of the longitudinal section of the (a) as-cast and (b) as-drawn Mn50Fe50 samples.
ARTICLE IN PRESS 3780
A. He et al. / Journal of Magnetism and Magnetic Materials 321 (2009) 3778–3781
Fig. 4. (2 0 0) and (2 2 0) pole figures measured on the as-drawn Mn50Fe50 alloy wire. DD denotes the drawing direction. Intensity contours in these pole figures are indicated by colors. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 5. Magnetostriction of the as-cast and as-drawn Mn50Fe50 samples measured along DD.
4. Discussion Mn–Fe alloys can exhibit as single g phase terminal solid solution alloys with a fcc structure for Mn contents in the range of 40–75 at% at room temperature, according to the Mn–Fe phase diagram [13]. In our previous work, it has been demonstrated that when Mn contents are below 40 at%, Mn–Fe alloys consisting of g and e phases possess much poor magnetostriction performance than the alloys with single g phase when Mn contents are above 40 at% [14]. It is also revealed that the magnetostrictive performance of Mn42Fe58 alloy deteriorates after annealing for 24 h at 1100 1C due to the separation of single g phase into a mixture of g, e and a phases [15]. These results suggest that single g phase in Mn–Fe alloys is essential to obtain large magnetostriction. In this work, the Mn50Fe50 alloy still remains a single g phase after hot rolling and cold drawing, ensuring the large magnetostriction. Plastic deformation occurs in metals by slip or by twinning on specific favorable atomic planes and crystallographic directions. In a polycrystalline sample, even though the initial orientation of grains is random, after sufficient deformation most of the grains
are realigned into a preferred orientation, hence developing a specific texture. From Fig. 4, it can be seen that the as-drawn Mn50Fe50 alloy wire shows a mixture of /1 0 0S and /11 0S fiber components. Combining with the microstructure observation (Fig. 3b), we put forward a primary explanation as follows. For the Mn50Fe50 alloy wire, a great amount of deformed grains are produced during rolling and drawing, and are arranged along DD. The nature of fiber textures in fcc metals mainly depends on the stacking fault energy (SFE) [16]. Jiang et al. have revealed that the fiber texture of the Fe–27Mn–6Si–5Cr alloy wire showed /1 0 0S and /11 0S duplex components [17]. The SFE of g phase in the Mn–Fe alloys with Mn atom ratio a (Mn) ¼ 20–50% could be calculated by embedded-atom method (EAM) as follows: SFE (mJ/m2) ¼ 23.3+0.269a (Mn)% [18]. According to this equation, the SFE in the g-Mn50Fe50 alloy is higher than that of g phase in the Fe–27Mn–6Si–5Cr alloy. Therefore, the /1 0 0S and /11 0S fiber textures parallel to DD is formed in the g-Mn50Fe50 alloy wire, which is the characteristic of fcc metals with a high SFE [19]. From Fig. 5, it can be seen that the magnetostriction of the drawn g-Mn50Fe50 alloy wire achieves 750 ppm under 1.2 T. The value is eleventh times as large as that of Fe–Ga wires [2], although it requires the use of two orders of magnitude at higher magnetic field. While the magnetostriction mechanism in antiferromagnetic g-Mn–Fe alloys remains unclear, owing to still lack in direct observation of antiferromagnetic domain structures in these alloys. However, the magnetostriction in an antiferromagnetic NiO crystal under an applied field, in principle, can be due to (i) the movement of twin boundaries, (ii) the alignment of antiferromagnetic domains, and (iii) the rotation of the spins [20]. For antiferromagnetic g-Mn50Fe50 alloy samples, there is no martensitic structure observed in the present work. That is to say, the absence of twin boundaries leads to a conclusion that the magnetostriction of the Mn50Fe50 alloy may be due to the movement of antiferromagnetism and/or the rotation of spins. Moreover, Peng and Zhang have predicted that the magnetostriction of Mn–Fe alloys could be affected by grain orientations and magnetic domain configurations [7]. Kennedy and Hicks have also demonstrated that the magnetic structure of the antiferromagnetic g-Mn–Fe alloys was a noncolinear spin structure with spin directions along /111S [21]. The magnetostriction of g-Mn–Fe alloys may be due to rotation of spins within a set of {111} planes. The mixture of /1 0 0S and /11 0S fiber components parallel to DD formed during deformation might result in the movement of domains and rotation of spins that becomes much easier during magnetizing
ARTICLE IN PRESS A. He et al. / Journal of Magnetism and Magnetic Materials 321 (2009) 3778–3781
process, which should be responsible for the magnetostrictive performance improvement. To further improve the magnetostriction of Mn–Fe alloys, single crystal growth might be a potential choice. Compared to the polycrystalline materials, improved magnetostriction has previously been achieved in single crystals, i.e. Terfenol-D and Ni–Mn–Ga alloys [5,22–24]. Saturation magnetostriction in polycrystalline Terfenol-D is just about 1000 ppm, but reaches 2375 ppm under a compressive pre-stress of 24 MPa and a magnetic field of 2.0 T in a single crystal prepared by Czochralski method. Meanwhile, single crystals of magnetic shape memory alloys Ni–Mn–Ga also possess much larger magnetostrain than the polycrystalline samples. For example, a giant magnetostrain of 6.3% can be achieved in a single crystal Ni50Mn27.5Ga22.5 with a five-layer (5M) structure, which is about one order of magnitude higher than the polycrystalline sample. Although it still lack in explanation of the magnetostrictive mechanism of antiferromagnetic Mn–Fe alloys, higher magnetostriction could be expected in single crystals. It is worth performing study on single crystal growth of Mn–Fe alloy in further work.
5. Conclusions An antiferromagnetic Mn50Fe50 alloy wire with large magnetostriction was successfully prepared by combining plastic deformation process through hot rolling and cold drawing. The retention of single g phase during deformation is essential to obtain large magnetostriction in the Mn50Fe50 alloy wire. A mixture of /1 0 0S and /11 0S duplex fiber texture along the drawing direction is formed in the as-drawn wire. Room temperature magnetostriction of up to 750 ppm under 1.2 T is achieved for the as-drawn Mn50Fe50 wire, which is twice as high as that of the as-cast state. Such large magnetostriction should be ascribed to the preferred crystal orientation along DD formed during plastic deformation.
3781
Acknowledgements This work was supported by the Natural Science Foundation of China (NSFC-50701039 and 50531010), Visiting Scholar Foundation of State Key Laboratory of Silicon Materials (SKL2009-6), and the Innovative Research Team in University (IRT-0651). References [1] K. Mohri, F.B. Humphrey, K. Kawashima, K. Kimura, M. Mizutani, IEEE Trans. Magn. 26 (1990) 1789. [2] J.H. Liu, F. Yi, C.B. Jiang, J. Alloys Compd. 481 (2009) 57. [3] C. Luna, V. Raposo, G. Rauscher, M. Va zquez, IEEE Trans. Magn. 38 (2002) 3108. [4] V. Raposoa, C. Lunaa, M. Va zquez, J. Magn. Magn. Mater. 249 (2002) 22. [5] A.E. Clark, Ferromagnetic Materials, Vol. 1, North-Holland, Amsterdam, 1980 P.531. [6] T.Y. Ma, M. Yan, J.J. Zhang, W. Luo, C.B. Jiang, H.B. Xu, Appl. Phys. Lett. 90 (2007) 102502. [7] W.Y. Peng, J.H. Zhang, Appl. Phys. Lett. 89 (2006) 262501. [8] S. Guruswamy, N. Srisukhumbowarnchai, A.E. Clark, J.B. Restorff, M. WunFogle, Scr. Mater. 43 (2000) 239. [9] J.H. Li, X.X. Gao, J. Zhu, C.X. He, J.W. Qiao, M.C. Zhao, J. Alloys Compd. 476 (2009) 529. [10] O. Grassela, L. Kruger, G. Frommeyer, L.W. Meyer, Int. J. Plast. 16 (2000) 1391. [11] A. Druker, C. Sobrero, H.G. Brokmeier, J. Malar\curr {r´}a, R. Bolmaro, Mater. Sci. Eng. A 481 (2008) 578. [12] T.Y. Ma, J.J. Zhang, A.N. He, M. Yan, Scr. Mater. 61 (2009) 427. [13] H.P. Kneijnsberg, C.A. Verbraak, M.J. Ten Bouwhuijs, Acta Metall. 33 (1985) 1759. [14] Y.W. Xu, T.Y. Ma, J.J. Zhang, M. Yan, Acta Metall. Sin. 44 (2008) 1235. [15] J.J. Zhang, T.Y. Ma, A.N. He, M. Yan, J. Alloys Compd. doi: 10.1016/j.jallcom. 2009.06.004. [16] A.T. English, G.Y. Chin, Acta Metall. 13 (1965) 1013. [17] B.H. Jiang, X. Qi, Y.L. Ren, C.Q. Wang, Mater. Trans. JIM 41 (2000) 663. [18] Y.H. Rong, Q.P. Meng, G. He, T.Y. Hsu, J. Shanghai Jiaotong Univ. 37 (2003) 171. [19] C.S. Lee, B.J. Duggan, Metall. Trans. A 22A (1991) 2637. [20] L. Alberts, E.W. Lee, Proc. Phys. Soc. London 78 (1961) 728. [21] K.J. Kennedy, T.J. Hicks, J. Phys. F: Met. Phys. 17 (1987) 1599. [22] G.H. Wu, X.G. Zhao, J.H. Wang, J.Y. Li, K.C. Jia, W.S. Zhan, Appl. Phys. Lett. 67 (1995) 2005. [23] R.C. O’Handley, J. Appl. Phys. 83 (1998) 3263. [24] C.B. Jiang, J.M. Wang, H.B. Xu, Appl. Phys. Lett. 86 (2005) 252508.