- Email: [email protected]
Magnetic properties of (1 1 0)- and (2 0 0)-oriented Fe-nanowire arrays
ARTICLE IN PRESS
Physica B 368 (2005) 100–104 www.elsevier.com/locate/physb
Magnetic properties of (1 1 0)- and (2 0 0)-oriented Fe-nanowire arrays H.N. Hu, H.Y. Chen, J.L. Chen, G.H. Wu Beijing National Laboratory for Condensed Matter Physics, Institute of Physics, Chinese Academy of Sciences, Beijing 100080, China Received 27 January 2005; received in revised form 4 July 2005; accepted 4 July 2005
Abstract Fe-nanowire arrays with (1 1 0) and (2 0 0) orientation have been fabricated through controlling the pH values in electrodeposition. Fe-nanowire arrays 30 and 60 nm in diameter were obtained. With the magnetic field applied parallel to the wire, 60 nm diameter Fe nanowires with preferred (2 0 0) orientation show an improved squareness and an easier magnetization than the nanowires with preferred (1 1 0) orientation. For 30 nm diameter Fe nanowires, (2 0 0)-oriented nanowires show an improved squareness and coercivity compared with (1 1 0)-oriented nanowires. r 2005 Elsevier B.V. All rights reserved. PACS: 75.75; 81.07; 81.07.B Keywords: Nanowire; Electrodeposition; Preferred orientation; Magnetic properties
1. Introduction One-dimensional nanostructures are of great interest because of their potential application in many areas, such as high-density perpendicularmagnetic-recording media and nanosensors [1,2]. The synthesis and precise control of such a magnetic nanostructure on a large scale is a challenging issue in material science. One strategy is to electrodeposit magnetic nanowires into nanochannels of porous Corresponding authors. Tel.:+86 10 8264 9247; fax:+86 10 8264 9485. E-mail addresses: [email protected], [email protected] (H.N. Hu).
anodic aluminum oxide (AAO) templates, which have been utilized by many groups to prepare magnetic metals, such as Ni, Co, and Fe. Among earlier works, most of the Fe nanowires deposited are (1 1 0) oriented or polycrystalline [3–7]. It is wellknown that (1 1 0) is the hard axis of the magnetization of Fe. Most of the earlier work was focused on ultra-thin Fe-nanowire arrays with diameters of about 5–35 nm [4–6], using the shape anisotropy to suppress the crystalline anisotropy. In the present paper, we report a unique dynamically controlled growth method to prepare Fe nanowire with preferred (2 0 0) orientation along the wire. In contrast, (1 1 0)-oriented Fe nanowires have also been deposited. Fe nanowires 60 and 30 nm diameter
0921-4526/$ - see front matter r 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.physb.2005.07.002
ARTICLE IN PRESS H.N. Hu et al. / Physica B 368 (2005) 100–104
2. Experimental procedures Porous AAO template was prepared by a twostep anodizing process on aluminum foils with a high purity of 99.999% [8,9]. Where AAO template with 60 nm diameter was anodized in 0.3 M oxalic acid solution under a constant voltage of 40 V at 12 1C, 30 nm diameter AAO template was anodized in 0.3 M sulfuric acid at 26 V and 1 1C. Both kinds of AAO templates were anodized for about 12 h. The 30 and 60 nm diameter templates have a thickness of about 50 and 70 mm, respectively. Both templates have a high aspect ratio of more than 1000. After anodization, the remaining aluminium substrate and its barrier layer at the bottom of the AAO template were removed. Then, a 300 nm Cu layer was sputterdeposited on one side of the AAO template to serve as the working electrode during the electrodeposition. An aqueous bath containing Fe2+ was used to deposit Fe nanowires at room temperature. Both kinds of Fe nanowires were electrodeposited using the potentiostatic method with a three-electrode arrangement with a saturated calomel reference electrode (SCE) and a graphite rod as counter electrode at a constant potential of 1.1 V. X-ray diffraction (XRD) with Cu Ka radiation was used to characterize the structure of the wires. Magnetic measurements were performed using a SQUID magnetometer. The electrodeposition was carried out as follows: (1) Fe-1 (60 nm diameter) and Fe-3 (30 nm diameter) were deposited in a solution containing 0.2 M FeCl2 with pH ¼ 2.6 (adjusted by an appropriate amount of dilute HCl). Then deposition was performed at a constant potential of
1.1 V relative to the SCE in a three-electrode system. (2) Fe-2 (60 nm diameter) and Fe-4 (30 nm diameter) were deposited in a solution containing 0.2 M FeCl2 with the same deposition potential of 1.1 V relative to SCE as Fe-1 and Fe-3. At the very beginning of the deposition procedure at pH ¼ 3.7, within 2 min the pH value was adjusted to 2.6 by adding an appropriate amount of dilute HCl; then the deposition was continued till Fe nanowire filled the whole AAO template. For this kind of sample, it was found that only changing of the pH value during the deposition procedure can result in a change of preferred orientation.
3. Results and discussion Fig. 1 shows the XRD spectra of Fe nanowire arrays with (1 1 0) (a) and (2 0 0) (b) orientation, embedded in the AAO template. Both samples show (1 1 0) and (2 0 0) reflections of cubic Fe. In Fig. 1(a) a very strong (1 1 0) reflection can be seen which is accompanied by two weak (2 0 0) and (2 1 1) reflections, indicating that the Fe-1 and Fe-3 nanowires have a strong preferred (1 1 0) orientation. Fig. 1(b) shows a very strong (2 0 0) reflection accompanied by a weak (1 1 0) reflection, indicat(a) Fe-1 Fe-3
[110]
8000
Intensity (cps)
with both orientations were fabricated. The magnetic properties were measured in a superconducting quantum interference device (SQUID) magnetometer. When the magnetic field is applied parallel to the wire, the 60 nm diameter Fe nanowires with (2 0 0) orientation show a higher aspect ratio and easier magnetization than the (1 1 0)-oriented Fe nanowires. As to the 30 nm Fe nanowire, (2 0 0)oriented nanowires show a much improved coercivity (about 30%) than (1 1 0)-oriented nanowires.
101
4000 [200]
[211]
0 [200]
[200] [110]
(b) Fe-2 Fe-4
4000
2000 [110]
60
80
0 40
60
80
2θ (deg.) Fig. 1. XRD of Fe nanowires electrodeposited under different conditions: (a) Fe-1 and Fe-3, (b) Fe-2 and Fe-4 (inside was XRD of the beginning growth side of the nanowire).
ARTICLE IN PRESS 102
H.N. Hu et al. / Physica B 368 (2005) 100–104
ing that Fe-2 and Fe-4 nanowires have a strong preferred (2 0 0) orientation. To make clear the growth mechanism of the two kinds of samples, the under-layer of Cu was removed and XRD was performed on the beginning growth side of Fe nanowires. For (1 1 0)oriented samples (Fe-1 and Fe-3), similar XRD spectra were obtained at both sides of the AAO template. For the (2 0 0)-oriented samples (Fe-2 and Fe-4), a strong (1 1 0) reflection with almost the same intensity as the (2 0 0) reflection can be seen at the beginning side of the samples (inset in Fig. 1(b)). This indicates that (2 0 0)-oriented samples also had preferred (1 1 0) orientation at the beginning of growth. Considering the change of pH value within 2 min from the start of the growth, the part of the sample with (1 1 0) orientation will be very small. Therefore, the Xrays can penetrate into the (2 0 0)-oriented part through the (1 1 0) part and the diagram exhibits a large (1 1 0) reflection with an equivalent (2 0 0) reflection. This implies that the change of growth orientation from (1 1 0) to (2 0 0) has occurred at the change of pH value from 3.7 to 2.6. Typical images of Fe-2 (a) and Fe-3 (b) obtained by scanning electron microscopy (SEM) are shown in Fig. 2. Both kinds of nanowires are continuous and have a uniform diameter that corresponds to the used AAO template. The selected-area electron diffraction (SEAD) in the inset of Fig. 2 shows that Fe-2 (a) nanowire has a preferred(2 0 0) orientation along the wire, whereas Fe-3 (b) nanowire has a preferred(1 1 0) orientation. SEAD also reveals the single-crystal structure of the two samples, although with some stacking faults and twinned crystals. Fig. 3 shows the hysteresis loops of Fe-1 and Fe2 (both with diameters of 60 nm) at 5 K. In a magnetic field applied parallel to the wire, Fe-1 nanowire arrays show a high saturation magnetization at about 7000 Oe with a low squareness of 14% and a coercivity of 630 Oe, shown in Fig. 3(a). For Fe-2 nanowire arrays, Fig. 3(b) shows a low saturation magnetization at about 3000 Oe with a squareness of 53% and a coercivity of 490 Oe. It is apparent that Fe-2 is much easier magnetized than Fe-1 nanowire. This may be understood in terms of the different preferred orientations of the two samples. Because (1 1 0) is
Fig. 2. TEM images and selected-area electron diffraction (inset of Fig. 2) of (a) Fe-2 and (b) Fe-3 single nanowires separated from the AAO template.
the hard magnetization direction of Fe and (2 0 0) the easy direction, Fe-1 with preferred (1 1 0) orientation will be much harder to saturate than Fe-2 with a preferred (2 0 0) orientation. It also can be seen clearly that both the (1 1 0)- and the (2 0 0)oriented nanowire arrays have a magnetic easy axis parallel to the wire. From a conventional point of view, the effective anisotropy of the magnetic nanowire arrays results from the competition between the magnetocrystalline anisotropy and the shape anisotropy. The cubic anisotropy constant of Fe is K1 ¼ 5 105 erg/cm3. The nanowire arrays with a high aspect ratio (up to 1000) can be considered to an infinite cylinder. Then the shape anisotropy at room temperature will be
ARTICLE IN PRESS
1.0
1.0
0.5
0.5 M/Ms
M/Ms
H.N. Hu et al. / Physica B 368 (2005) 100–104
0.0
(a) Fe-1
-1.0 1.0
(a) Fe-3
-1.0 1.0 0.5 M/Ms
0.5 M/Ms
0.0 -0.5
-0.5
0.0
0.0 -0.5
-0.5 (b) Fe-2
-1.0 -10000
103
-5000
0
5000
10000
(b) Fe-4
-1.0 -10000
-5000
0
5000
10000
H(Oe)
H(Oe) Fig. 3. Hysteresis loops of (a) Fe-1 and (b) Fe-2 (60 nm diameter) at 5 K. (——) denotes field applied parallel to the wire and (—J—) denotes field applied perpendicular to the wire.
Fig. 4. Hysteresis loops of (a) Fe-3 and (b) Fe-4 (30 nm diameter) at 5 K. (——) denotes field applied parallel to the wire and (—J—) denotes field applied perpendicular to the wire.
pM 2S ¼ 9:5 106 erg=cm3 (where M S ¼ 217 Oe), an order of magnitude larger than the crystallographic anisotropy [10]. Thus both the anisotropy of Fe-1 (with (1 1 0) orientation) and the anisotropy of Fe-2 (with (2 0 0) orientation) show a magnetic easy axis parallel to the wire. The hysteresis loops of Fe-3 (a) and Fe-4 (b) (30 nm diameter) were also measured at 5 K (Fig. 4). It can be seen in Fig. 4(a) that Fe-3 nanowire arrays have a coercivity of 1690 Oe and an aspect ratio of 95% and that Fe-4 nanowire arrays have a coercivity of 2270 Oe and an aspect ratio of 98.4%. Both samples show a much improved coercivity and squareness than the 60 nm diameter nanowires. With a decrease of the diameter, the shape anisotropy enhances and then exceeds the crystalline anisotropy, as has also been reported by Yang and Zhang [4,5]. Comparing Fe-3 with Fe-4, the Fe-4 nanowire arrays with preferred (2 0 0) orientation have a coercivity which is about 580 Oe larger than Fe-3 arrays (about
30% higher). This indicates that, in case of the same diameter, the crystalline anisotropy will induce a preferred easy magnetization direction. This changing of the pH value can be used to make Fe nanowires with even smaller diameter, and it will be certain that the nanowire arrays with (2 0 0) preferred orientation and with smaller diameter will show even higher coercivity.
4. Conclusion In conclusion, a unique method in which the pH value is dynamically changed was used to fabricate Fe-nanowire arrays with preferred (1 1 0) and (2 0 0) orientation. Fe-nanowire arrays with 30 and 60 nm diameter were obtained in AAO templates. With the magnetic field applied parallel to the wire, 60 nm diameter Fe nanowires with preferred (2 0 0) orientation show an improved squareness and are easier
ARTICLE IN PRESS 104
H.N. Hu et al. / Physica B 368 (2005) 100–104
magnetized than nanowire with preferred (1 1 0) orientation. For 30 nm Fe nanowires with preferred (2 0 0) orientation, a higher coercivity than wires with preferred (1 1 0) orientation was found. This is of much interest for high-density perpendicular-magnetic-recording media and nanosensors. This is the first time that preferred (2 0 0)-oriented Fe nanowires have been reported with crystalline anisotropy and shape anisotropy in the same direction along the wire. Apparently, the dynamic change of the pH value induces the change of preferred (1 1 0)-oriented to (2 0 0)-oriented nanowire. The origin of this influence of the pH value upon the growth direction needs further investigation.
Acknowledgements This work was supported by National Natural Science Foundation of China Grant No. 50371101.
Reference [1] D. Almawlawi, N. Coombs, M. Moskovits, J. Appl. Phys. 70 (1991) 4421. [2] T. Thurn-Albrecht, J. Schotter, G.A. Kastle, N. Emley, T. Shibauchi, L. Krusin-Elbaum, K. Guarini, C.T. Black, M.T. Tuominen, T.P. Russell, Science 290 (2000) 2126. [3] V. Langlais, S. Arrii, L. Pontonnier, G. Tourillon, Scripta Mater 44 (2001) 1315. [4] S.G. Yang, H. Zhu, D.L. Yu, Z.Q. Jin, S.L. Tang, Y.W. Du, J. Magn. Magn. Mater. 222 (2000) 97. [5] X.Y. Zhang, G..H. Wen, Y.F. Chan, R.K. Zheng, X.X. Zhang, N. Wang, Appl. Phys. Lett. 83 (2003) 3341. [6] D.J. Sellmyer, M. Zheng, R. Skomski, J. Phys.: Condens. Matter 13 (2001) R433. [7] F.S. Li, L.Y. Ren, Z.P. Niu, H.X. Wang, T. Wang, J. Phys.: Condens. Matter 14 (2002) 6875. [8] H. Masuda, K. Fukuda, Science 268 (1995) 1466. [9] Y.W. Wang, L.D. Zhang, G.W. Meng, X.S. Peng, Y.X. Jin, J. Zhang, J. Phys. Chem. B 106 (2002) 2502. [10] W.D. Zhong, Ferromagnetism, vol. 2, Ch. 7 (Science Press, 1998, in Chinese).
Physica B 368 (2005) 100–104 www.elsevier.com/locate/physb
Magnetic properties of (1 1 0)- and (2 0 0)-oriented Fe-nanowire arrays H.N. Hu, H.Y. Chen, J.L. Chen, G.H. Wu Beijing National Laboratory for Condensed Matter Physics, Institute of Physics, Chinese Academy of Sciences, Beijing 100080, China Received 27 January 2005; received in revised form 4 July 2005; accepted 4 July 2005
Abstract Fe-nanowire arrays with (1 1 0) and (2 0 0) orientation have been fabricated through controlling the pH values in electrodeposition. Fe-nanowire arrays 30 and 60 nm in diameter were obtained. With the magnetic field applied parallel to the wire, 60 nm diameter Fe nanowires with preferred (2 0 0) orientation show an improved squareness and an easier magnetization than the nanowires with preferred (1 1 0) orientation. For 30 nm diameter Fe nanowires, (2 0 0)-oriented nanowires show an improved squareness and coercivity compared with (1 1 0)-oriented nanowires. r 2005 Elsevier B.V. All rights reserved. PACS: 75.75; 81.07; 81.07.B Keywords: Nanowire; Electrodeposition; Preferred orientation; Magnetic properties
1. Introduction One-dimensional nanostructures are of great interest because of their potential application in many areas, such as high-density perpendicularmagnetic-recording media and nanosensors [1,2]. The synthesis and precise control of such a magnetic nanostructure on a large scale is a challenging issue in material science. One strategy is to electrodeposit magnetic nanowires into nanochannels of porous Corresponding authors. Tel.:+86 10 8264 9247; fax:+86 10 8264 9485. E-mail addresses: [email protected], [email protected] (H.N. Hu).
anodic aluminum oxide (AAO) templates, which have been utilized by many groups to prepare magnetic metals, such as Ni, Co, and Fe. Among earlier works, most of the Fe nanowires deposited are (1 1 0) oriented or polycrystalline [3–7]. It is wellknown that (1 1 0) is the hard axis of the magnetization of Fe. Most of the earlier work was focused on ultra-thin Fe-nanowire arrays with diameters of about 5–35 nm [4–6], using the shape anisotropy to suppress the crystalline anisotropy. In the present paper, we report a unique dynamically controlled growth method to prepare Fe nanowire with preferred (2 0 0) orientation along the wire. In contrast, (1 1 0)-oriented Fe nanowires have also been deposited. Fe nanowires 60 and 30 nm diameter
0921-4526/$ - see front matter r 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.physb.2005.07.002
ARTICLE IN PRESS H.N. Hu et al. / Physica B 368 (2005) 100–104
2. Experimental procedures Porous AAO template was prepared by a twostep anodizing process on aluminum foils with a high purity of 99.999% [8,9]. Where AAO template with 60 nm diameter was anodized in 0.3 M oxalic acid solution under a constant voltage of 40 V at 12 1C, 30 nm diameter AAO template was anodized in 0.3 M sulfuric acid at 26 V and 1 1C. Both kinds of AAO templates were anodized for about 12 h. The 30 and 60 nm diameter templates have a thickness of about 50 and 70 mm, respectively. Both templates have a high aspect ratio of more than 1000. After anodization, the remaining aluminium substrate and its barrier layer at the bottom of the AAO template were removed. Then, a 300 nm Cu layer was sputterdeposited on one side of the AAO template to serve as the working electrode during the electrodeposition. An aqueous bath containing Fe2+ was used to deposit Fe nanowires at room temperature. Both kinds of Fe nanowires were electrodeposited using the potentiostatic method with a three-electrode arrangement with a saturated calomel reference electrode (SCE) and a graphite rod as counter electrode at a constant potential of 1.1 V. X-ray diffraction (XRD) with Cu Ka radiation was used to characterize the structure of the wires. Magnetic measurements were performed using a SQUID magnetometer. The electrodeposition was carried out as follows: (1) Fe-1 (60 nm diameter) and Fe-3 (30 nm diameter) were deposited in a solution containing 0.2 M FeCl2 with pH ¼ 2.6 (adjusted by an appropriate amount of dilute HCl). Then deposition was performed at a constant potential of
1.1 V relative to the SCE in a three-electrode system. (2) Fe-2 (60 nm diameter) and Fe-4 (30 nm diameter) were deposited in a solution containing 0.2 M FeCl2 with the same deposition potential of 1.1 V relative to SCE as Fe-1 and Fe-3. At the very beginning of the deposition procedure at pH ¼ 3.7, within 2 min the pH value was adjusted to 2.6 by adding an appropriate amount of dilute HCl; then the deposition was continued till Fe nanowire filled the whole AAO template. For this kind of sample, it was found that only changing of the pH value during the deposition procedure can result in a change of preferred orientation.
3. Results and discussion Fig. 1 shows the XRD spectra of Fe nanowire arrays with (1 1 0) (a) and (2 0 0) (b) orientation, embedded in the AAO template. Both samples show (1 1 0) and (2 0 0) reflections of cubic Fe. In Fig. 1(a) a very strong (1 1 0) reflection can be seen which is accompanied by two weak (2 0 0) and (2 1 1) reflections, indicating that the Fe-1 and Fe-3 nanowires have a strong preferred (1 1 0) orientation. Fig. 1(b) shows a very strong (2 0 0) reflection accompanied by a weak (1 1 0) reflection, indicat(a) Fe-1 Fe-3
[110]
8000
Intensity (cps)
with both orientations were fabricated. The magnetic properties were measured in a superconducting quantum interference device (SQUID) magnetometer. When the magnetic field is applied parallel to the wire, the 60 nm diameter Fe nanowires with (2 0 0) orientation show a higher aspect ratio and easier magnetization than the (1 1 0)-oriented Fe nanowires. As to the 30 nm Fe nanowire, (2 0 0)oriented nanowires show a much improved coercivity (about 30%) than (1 1 0)-oriented nanowires.
101
4000 [200]
[211]
0 [200]
[200] [110]
(b) Fe-2 Fe-4
4000
2000 [110]
60
80
0 40
60
80
2θ (deg.) Fig. 1. XRD of Fe nanowires electrodeposited under different conditions: (a) Fe-1 and Fe-3, (b) Fe-2 and Fe-4 (inside was XRD of the beginning growth side of the nanowire).
ARTICLE IN PRESS 102
H.N. Hu et al. / Physica B 368 (2005) 100–104
ing that Fe-2 and Fe-4 nanowires have a strong preferred (2 0 0) orientation. To make clear the growth mechanism of the two kinds of samples, the under-layer of Cu was removed and XRD was performed on the beginning growth side of Fe nanowires. For (1 1 0)oriented samples (Fe-1 and Fe-3), similar XRD spectra were obtained at both sides of the AAO template. For the (2 0 0)-oriented samples (Fe-2 and Fe-4), a strong (1 1 0) reflection with almost the same intensity as the (2 0 0) reflection can be seen at the beginning side of the samples (inset in Fig. 1(b)). This indicates that (2 0 0)-oriented samples also had preferred (1 1 0) orientation at the beginning of growth. Considering the change of pH value within 2 min from the start of the growth, the part of the sample with (1 1 0) orientation will be very small. Therefore, the Xrays can penetrate into the (2 0 0)-oriented part through the (1 1 0) part and the diagram exhibits a large (1 1 0) reflection with an equivalent (2 0 0) reflection. This implies that the change of growth orientation from (1 1 0) to (2 0 0) has occurred at the change of pH value from 3.7 to 2.6. Typical images of Fe-2 (a) and Fe-3 (b) obtained by scanning electron microscopy (SEM) are shown in Fig. 2. Both kinds of nanowires are continuous and have a uniform diameter that corresponds to the used AAO template. The selected-area electron diffraction (SEAD) in the inset of Fig. 2 shows that Fe-2 (a) nanowire has a preferred(2 0 0) orientation along the wire, whereas Fe-3 (b) nanowire has a preferred(1 1 0) orientation. SEAD also reveals the single-crystal structure of the two samples, although with some stacking faults and twinned crystals. Fig. 3 shows the hysteresis loops of Fe-1 and Fe2 (both with diameters of 60 nm) at 5 K. In a magnetic field applied parallel to the wire, Fe-1 nanowire arrays show a high saturation magnetization at about 7000 Oe with a low squareness of 14% and a coercivity of 630 Oe, shown in Fig. 3(a). For Fe-2 nanowire arrays, Fig. 3(b) shows a low saturation magnetization at about 3000 Oe with a squareness of 53% and a coercivity of 490 Oe. It is apparent that Fe-2 is much easier magnetized than Fe-1 nanowire. This may be understood in terms of the different preferred orientations of the two samples. Because (1 1 0) is
Fig. 2. TEM images and selected-area electron diffraction (inset of Fig. 2) of (a) Fe-2 and (b) Fe-3 single nanowires separated from the AAO template.
the hard magnetization direction of Fe and (2 0 0) the easy direction, Fe-1 with preferred (1 1 0) orientation will be much harder to saturate than Fe-2 with a preferred (2 0 0) orientation. It also can be seen clearly that both the (1 1 0)- and the (2 0 0)oriented nanowire arrays have a magnetic easy axis parallel to the wire. From a conventional point of view, the effective anisotropy of the magnetic nanowire arrays results from the competition between the magnetocrystalline anisotropy and the shape anisotropy. The cubic anisotropy constant of Fe is K1 ¼ 5 105 erg/cm3. The nanowire arrays with a high aspect ratio (up to 1000) can be considered to an infinite cylinder. Then the shape anisotropy at room temperature will be
ARTICLE IN PRESS
1.0
1.0
0.5
0.5 M/Ms
M/Ms
H.N. Hu et al. / Physica B 368 (2005) 100–104
0.0
(a) Fe-1
-1.0 1.0
(a) Fe-3
-1.0 1.0 0.5 M/Ms
0.5 M/Ms
0.0 -0.5
-0.5
0.0
0.0 -0.5
-0.5 (b) Fe-2
-1.0 -10000
103
-5000
0
5000
10000
(b) Fe-4
-1.0 -10000
-5000
0
5000
10000
H(Oe)
H(Oe) Fig. 3. Hysteresis loops of (a) Fe-1 and (b) Fe-2 (60 nm diameter) at 5 K. (——) denotes field applied parallel to the wire and (—J—) denotes field applied perpendicular to the wire.
Fig. 4. Hysteresis loops of (a) Fe-3 and (b) Fe-4 (30 nm diameter) at 5 K. (——) denotes field applied parallel to the wire and (—J—) denotes field applied perpendicular to the wire.
pM 2S ¼ 9:5 106 erg=cm3 (where M S ¼ 217 Oe), an order of magnitude larger than the crystallographic anisotropy [10]. Thus both the anisotropy of Fe-1 (with (1 1 0) orientation) and the anisotropy of Fe-2 (with (2 0 0) orientation) show a magnetic easy axis parallel to the wire. The hysteresis loops of Fe-3 (a) and Fe-4 (b) (30 nm diameter) were also measured at 5 K (Fig. 4). It can be seen in Fig. 4(a) that Fe-3 nanowire arrays have a coercivity of 1690 Oe and an aspect ratio of 95% and that Fe-4 nanowire arrays have a coercivity of 2270 Oe and an aspect ratio of 98.4%. Both samples show a much improved coercivity and squareness than the 60 nm diameter nanowires. With a decrease of the diameter, the shape anisotropy enhances and then exceeds the crystalline anisotropy, as has also been reported by Yang and Zhang [4,5]. Comparing Fe-3 with Fe-4, the Fe-4 nanowire arrays with preferred (2 0 0) orientation have a coercivity which is about 580 Oe larger than Fe-3 arrays (about
30% higher). This indicates that, in case of the same diameter, the crystalline anisotropy will induce a preferred easy magnetization direction. This changing of the pH value can be used to make Fe nanowires with even smaller diameter, and it will be certain that the nanowire arrays with (2 0 0) preferred orientation and with smaller diameter will show even higher coercivity.
4. Conclusion In conclusion, a unique method in which the pH value is dynamically changed was used to fabricate Fe-nanowire arrays with preferred (1 1 0) and (2 0 0) orientation. Fe-nanowire arrays with 30 and 60 nm diameter were obtained in AAO templates. With the magnetic field applied parallel to the wire, 60 nm diameter Fe nanowires with preferred (2 0 0) orientation show an improved squareness and are easier
ARTICLE IN PRESS 104
H.N. Hu et al. / Physica B 368 (2005) 100–104
magnetized than nanowire with preferred (1 1 0) orientation. For 30 nm Fe nanowires with preferred (2 0 0) orientation, a higher coercivity than wires with preferred (1 1 0) orientation was found. This is of much interest for high-density perpendicular-magnetic-recording media and nanosensors. This is the first time that preferred (2 0 0)-oriented Fe nanowires have been reported with crystalline anisotropy and shape anisotropy in the same direction along the wire. Apparently, the dynamic change of the pH value induces the change of preferred (1 1 0)-oriented to (2 0 0)-oriented nanowire. The origin of this influence of the pH value upon the growth direction needs further investigation.
Acknowledgements This work was supported by National Natural Science Foundation of China Grant No. 50371101.
Reference [1] D. Almawlawi, N. Coombs, M. Moskovits, J. Appl. Phys. 70 (1991) 4421. [2] T. Thurn-Albrecht, J. Schotter, G.A. Kastle, N. Emley, T. Shibauchi, L. Krusin-Elbaum, K. Guarini, C.T. Black, M.T. Tuominen, T.P. Russell, Science 290 (2000) 2126. [3] V. Langlais, S. Arrii, L. Pontonnier, G. Tourillon, Scripta Mater 44 (2001) 1315. [4] S.G. Yang, H. Zhu, D.L. Yu, Z.Q. Jin, S.L. Tang, Y.W. Du, J. Magn. Magn. Mater. 222 (2000) 97. [5] X.Y. Zhang, G..H. Wen, Y.F. Chan, R.K. Zheng, X.X. Zhang, N. Wang, Appl. Phys. Lett. 83 (2003) 3341. [6] D.J. Sellmyer, M. Zheng, R. Skomski, J. Phys.: Condens. Matter 13 (2001) R433. [7] F.S. Li, L.Y. Ren, Z.P. Niu, H.X. Wang, T. Wang, J. Phys.: Condens. Matter 14 (2002) 6875. [8] H. Masuda, K. Fukuda, Science 268 (1995) 1466. [9] Y.W. Wang, L.D. Zhang, G.W. Meng, X.S. Peng, Y.X. Jin, J. Zhang, J. Phys. Chem. B 106 (2002) 2502. [10] W.D. Zhong, Ferromagnetism, vol. 2, Ch. 7 (Science Press, 1998, in Chinese).