IEEE TRANSACTIONS ON MAGNETICS, VOL. 45, NO. 10, OCTOBER 2009
Magnetotransport Phenomena in [NiFe/Cu] Magnetic Multilayered Nanowires Horia Chiriac, Oana-Georgiana Dragos, Marian Grigoras, Gabriel Ababei, and Nicoleta Lupu National Institute of Research and Development for Technical Physics, Iasi RO-700050, Romania [Py(20–50 nm)/Cu(5–20 nm)] x n multilayered nanowires were electrodeposited by switching between the deposition potentials of the two constituents [respectively, 1 4 V and 0 3 V for permalloy (NiFe) and Cu deposition]. The magnetotransport properties of single multilayered nanowires have been measured. The MR curve is symmetrically anhysteretic, independent of the thickness of the nonmagnetic Cu layer. The MR ratio can reach values of 2%–2.5%, depending on the thickness of the nonmagnetic Cu layer. The presence of nonmagnetic Cu layers produces a decrease of the axial magnetic anisotropy in the multilayered nanowires, and consequently the magnetic permeability increases at the surface of the nanowires leading to the enhancement of the dc magnetic field effect over the MI response, which can go up to 100%. Index Terms—Giant magnetoimpedance, magnetic nanowires, magnetoresistance, multilayered structures.
I. INTRODUCTION AGNETIC nanowires were extensively studied in the last decade in order to understand and clarify a number of fundamental aspects or for potential applications –. A special emphasis is laid on the nanowires prepared by electrochemical deposition in the nanopores of different templates, due to the efficiency of this preparation method –. Additionally, the magnetotransport properties (both dc and ac) are behaving in a very specific manner in the multilayered nanowires structures, recommending them for applications in spintronics , , –. On the other hand, the high-frequency properties (mainly the magnetoimpedance) are extremely important for developing new systems of microsensors arrays, to be used both in technical and biomedical applications . The aim of this paper is to present some of our latest results on the magnetoresistance (MR) and magnetoimpedance (MI) effects in [Py/Cu] x n multilayered nanowires prepared by electrodeposition into the nanopores of the homemade aluminium anodized (AAO) templates.
II. EXPERIMENTAL PROCEDURE The anodic aluminum oxide template (AAO) was prepared by the two-step anodization process at 15 C in an oxalic acid solution 0.3 M. The anodization potential was 40 V. The process consists of two distinct steps: in the first one, the high purity (99.9999%) aluminum foil is subjected to the first anodization for 24 h; after the selective removal of the alumina film formed on the surface of the Al anodized foil with an aqueous solution of 1.8% H CrO and 6% H PO , the Al foil is subjected to the second anodization step for 19 h, the growth rate of the oxide layer being 6 m/h. After the selective removal of the aluminum substrate (with HgCl saturate solution) and the pore widening (with 5% H PO aqueous solution), the obtained nanopores have nominal diameters of 35 nm and are arranged hexagonally, Manuscript received March 06, 2009. Current version published September 18, 2009. Corresponding author: H. Chiriac (e-mail: [email protected]
). Digital Object Identifier 10.1109/TMAG.2009.2024636
Fig. 1. Top-view SEM image of a highly ordered AAO template.
as shown in Fig. 1. The density of the nanopores is about pores/cm and their length 50 m. A thin layer of Au and Cr was deposited by e-beam evaporation on one side of the AAO template, serving as working electrode for subsequent electrodeposition. A three electrode cell with a platinum wire as counter electrode and an Ag/AgCl reference electrode was used to carry out the electrodeposition process. Multilayered structures of magnetic permalloy (Ni Fe ) and nonmagnetic Cu were electrodeposited into the AAO template nanopores. At least 100 sequences were electrodeposited consecutively, using a single bath consisting of an aqueous solution of NiSO 90 g/L, FeSO 13.5 g/L, CuSO 2 g/L, and H BO 25 g/L. The pH value of the electrolyte was adjusted to 3 by adding few drops of 0.5 M H SO solution or 1 M NaOH solution. Multilayered structures Py(20, 30 or 50 nm)/Cu(5, 10 or 20 nm) were electrodeposited by switching between the deposition potentials of the two constituents (reV and V for Py and Cu deposition). The spectively, applied voltage was controlled with a HEKA PG 340 bipotentiostat/galvanostat. The existence of permalloy and Cu layers was evidenced by X-ray diffraction (BRUKER D8-Advance) measurements as well as by scanning electron microscopy (SEM) combined with energy dispersive spectroscopy (EDS). The SEM images were taken with a CrossBeam Neon 40 EsB microscope from Carl Zeiss SMT AG, whereas the EDS analysis was performed on a JEOL JSM 6390A. The magnetic hysteresis curves of single nanowires were measured by using a Kerr Magnetometer NanoMOKE2 from
0018-9464/$26.00 © 2009 IEEE
IEEE TRANSACTIONS ON MAGNETICS, VOL. 45, NO. 10, OCTOBER 2009
Durham Magneto Optics Ltd. For both magnetic measurements and EDS analysis, the AAO template was dissolved completely in a solution of NaOH 2.5 M. In the process used to contact a single nanowire for magnetotransport measurements, the whole assembly (AAO template—nanowires) is mechanically polished, using colloidal dim) and Syton (particles size amond (particles size nm), to have the nanowires ending at the template surface. When all nanowires are visible on the AAO surface, a thin layer (200 nm) of S1805 positive resist is spin-coated , . After the heat treatment at 150 C for 30 min (to tailor the mechanical properties of the resist layer), a second layer of S1805 positive resist is spin-coated and heated subsequently at 80 C for 30 min. In this second layer, a small square shape is defined by photolithography. At this stage, the sample is ready to be used for nanoindentation, which is performed by means of a Park XE-100 atomic force microscope in contact mode and using a conductive tip (I-AFM mode). The conductive tip scans the surface in contact mode and maps the surface sample topography. At the same time, the current flowing between the AFM conductive tip and the sample is measured to map the conductivity of the sample surface. Then, the indented hole is filled with a 300 nm Ag layer by thermal evaporation to establish contacts on single nanowires. After nanoindentation, the template is mounted on a sample holder with silver paste to ensure a good electrical conductivity between the bottom side of the indented nanowire and the holder. Finally, a gold contact wire is soldered on top of the silver layer filling the indented hole. All magnetotransport measurements were obtained at room temperature. The magnetoresistance of isolated spin valve nanowires was measured by four-probe method, as a function of the direction and amplitude of the external field. The maximum applied field was 8 kOe. The magnetoimpedance response was measured using a short-circuit coaxial cell connected at the VNA (Vector Network Analyzer Agilent N5230A) test port by the microwave flexible cable. The cell has the shape of a coaxial transmission line composed of a central pin with the external diameter of 3 mm, centrally positioned and electrically isolated in a metallic cylinder with the inner diameter of 7 mm. The central pin has one sharp end with the radius below 100 m, which is used to make the electrical contact with the silver thin layer deposited on the indented nanowires template. A SOL (Short-Open-Load) calibration kit was used to calibrate the measuring cell. The cell is inserted between the poles of an electromagnet which creates kOe, parallel to the a maximum dc magnetic field nanowires axis. The microwave signal frequency was ranging from 500 MHz to 18 GHz. For the calculation of the relative variation of the impedance of the single contact nanowire, the following relation was used:
% where Z(H) is the sample’s impedance at a certain value of the is the sample’s impedance at the magnetic field and maximum value of the magnetic field.
Fig. 2. HRSEM image of [Py(50 nm)/Cu(10 nm)]
2 n multilayered nanowires.
Fig. 3. MR curves of [Py(50 nm)/Cu(5 nm)] n and [Py(50 nm)/Cu(10 nm)] x n single contacted multilayered nanowires (8 = 35 nm) with the magnetic field in the plane of the layers.
III. RESULTS AND DISCUSSION Fig. 2 presents the high-resolution SEM image of multilayered nanowires of [Py(50 nm)/Cu(10 nm)] n, after dissolution of AAO template. Fig. 3 presents the magnetoresistance (MR) variation with the applied magnetic field for single contacted multilayered nanowires with the thickness of the Cu non-magnetic layer of 5 and 10 nm, respectively. The applied field was perpendicular on the nanowire’s growth direction, i.e., in-plane of the successive deposited layers of Py and Cu. The current passing through the nanowire was 5 A. Despite the fact that the electrical contact was made on a single nanowire, the variation of the magnetoresistance is reaching significant values (around 2%), comparable with those already reported in the literature for similar single contacted nanowires with larger diameters . The MR curve is symmetrically anhysteretic, independent of the thickness of the nonmagnetic Cu layer. However, the MR curve is not ideally square, as usually observed for GMR multilayered structures prepared by sputtering . The MR ratio can by modified by tailoring the thickness of the nonmagnetic Cu layer between the exchange interaction length and the spin-diffusion length .
CHIRIAC et al.: MAGNETOTRANSPORT PHENOMENA IN [NIFE/CU] MAGNETIC MULTILAYERED NANOWIRES
Fig. 4. GMI variation as a function of dc magnetic field and ac current frequency (f ) for [Py(30 nm)/Cu(10 nm)] n multilayered nanowires (8 = 35 nm), with the magnetic field out-of-plane of the layer.
Our MR measurements proved that the optimum Cu layer thickness is around 10 nm. When the external magnetic field is parallel with the nanowire (i.e., out-of-plane of the layer), the MR ratio decreases drastically, reaching only 0.10%–0.20%. The MI measurements were done by passing an ac current through the nanowire. The giant magnetoimpedance (GMI) response is reaching almost 100% for NiFe layers of 30 nm alternating with Cu layers of 10 nm (Fig. 4). The maximum GMI response is achieved at frequencies, , around 10 GHz and for dc applied magnetic fields of 300 Oe. The GMI response is very small for frequencies below 8 GHz and above 12 GHz. However, it is noted that in a narrow frequency range (8–12 GHz), the GMI response increases first from 55% to 95% with the frequency increase from 6 to 8 GHz, then reduces almost twice (around 40%) for a further increase of the frequency to 12 GHz. The variation of the frequency affects both the magnitude of the GMI response and the value of the dc magnetic field at which the maximum impedance response is achieved. The direction of the dc magnetic field relative to the layers composing the multilayered structure influences strongly the GMI response, as shown in Fig. 5. Additionally, the increase of the thickness of the ferromagnetic permalloy layer (the nonmagnetic Cu layer thickness remains constant) leads to the decrease of the GMI response to half or more, the effect being more pronounced when the dc magnetic field is in the plane of the layers. Both the magnitude of the GMI response and the dc magnetic field at which the GMI reaches its maximum are strongly dependent on the direction of the applied dc field, but also on the frequency of the ac current passing through the multilayered nanowire. Fig. 5 shows only the GMI curves at those frequencies for which the maximum is reached. The MI measurements done on permalloy nanowires with the same diameters show a very reduced variation (a few percents only) compared with multilayered [Py/Cu] x n nanowires, which indicates that the large GMI response is produced by the alternating ferromagnetic permalloy layers and nonmagnetic Cu layers. The presence of nonmagnetic Cu layers produces a decrease of the axial magnetic anisotropy in the multilayered nanowire,
Fig. 5. GMI variation as a function of dc magnetic field and ac current frequency (f ) for [Py(t nm)/Cu(10 nm)] n multilayered nanowires (t = 30 and 50; 8 = 35 nm), with the magnetic field in-the-plane and out-of-plane of the layer, respectively.
Fig. 6. Magneto-optical Kerr signal versus dc magnetic field for single nanowires of Py and [Py/Cu] n, respectively.
and consequently the circumferential magnetic permeability increases at the surface of the nanowires leading to the enhancement of the dc magnetic field effect over the MI response. Such a specific behavior is confirmed by the m-H loops obtained by magneto-optical Kerr effect measurements on the surface of single permaloy and [Py/Cu] x n multilayered nanowires, respectively (Fig. 6). The decrease of the coercive field for the individual multilayered nanowires compared with the simple permalloy nanowires is the result of the reduced axial magnetic anisotropy in the first case and the increase of the circumferential magnetic permeability in the surface layer, which is responsible for the GMI effect .
IEEE TRANSACTIONS ON MAGNETICS, VOL. 45, NO. 10, OCTOBER 2009
It is worthwhile to note that an important contribution to the GMI response is coming from the neighbor nanowires as well as from the dielectric alumina membrane. The quantification of these contributions requires further measurements, including different diameters and densities of nanowires, which will be the subject of our future investigations. IV. CONCLUSION [Py(20–50 nm)/Cu(5–20 nm)] n multilayered nanowires were electrodeposited by switching between the deposition V and potentials of the two constituents [respectively, V for permaloy (NiFe) and Cu deposition]. The magnetotransport properties of single multilayered nanowires have been measured. The single contacts have been made by a complex procedure, implying the deposition of a photoresist and nanoindentation using an AFM conductive tip. The MR curve is symmetrically anhysteretic, independent of the thickness of the nonmagnetic Cu layer. The MR ratio can by modified by tailoring the thickness of the nonmagnetic Cu layer between the exchange interaction length and the spin-diffusion length. The presence of nonmagnetic Cu layers produces a decrease of the axial magnetic anisotropy in the multilayered nanowires, and consequently the magnetic permeability increases at the surface of the nanowires leading to the enhancement of the dc magnetic field effect over the MI response. The results reported in the present work open up the possibility of realization of nanosensors arrays with enhanced sensitivity in a narrow range of frequencies. ACKNOWLEDGMENT This work was supported in part by the Romanian PN II—Partnerships Programme (Project NANOBIODET, Contract No. 11-072/2007) and in part by the European Community under the Sixth Framework Programme for the Marie Curie Research Training Network “SPINSWITCH” Contract Number MRTN-CT-2006-035327. REFERENCES  R. Cowburn and D. Petit, “Spintronics—Turbulence ahead,” Nature Mater., vol. 4, pp. 721–722, Oct. 2005.
 A. Fert and L. Piraux, “Magnetic nanowires,” J. Magn. Magn. Mater., vol. 200, pp. 338–358, 1999.  J.-E. Wegrowe, D. Kelly, A. Franck, S. E. Gilbert, and J.-Ph. Ansermet, “Magnetoresistance of ferromagnetic nanowires,” Phys. Rev. Lett., vol. 82, pp. 3681–3684, May 1999.  L. Piraux, K. Renard, R. Guillemet, S. Mátéfi-Templi, M. Mátéfi-Templi, V. A. Antohe, S. Fusil, K. Bouzehouane, and V. Cros, “Template-grown NiFe/Cu/NiFe nanowires for spin transfer devices,” Nano Lett., vol. 7, pp. 2563–2567, 2007.  K. Nielsch, F. Müller, A.-P. Li, and U. Gösele, “Uniform Ni deposition into ordered alumina pores by pulsed electrodeposition,” Adv. Mater., vol. 12, pp. 582–586, 2000.  R. M. Metzger, V. V. Konovalov, M. Sun, T. Xu, G. Zangari, B. Xu, M. Benakli, and W. D. Doyle, “Magnetic nanowires in hexagonally ordered pores of alumina,” IEEE Trans. Magn., vol. 36, pp. 30–35, Jan. 2000.  L.-P. Carignan, C. Lacroix, A. Ouimet, M. Ciureanu, A. Yelon, and D. Ménard, “Magnetic anisotropy in arrays of Ni, CoFeB, and Ni/Cu nanowires,” J. Appl. Phys., vol. 102, 2007, Art. No. 023905.  H. Chiriac, T. A. Óvári, and P. Pascariu, “Phenomenological model for the simulation of hysteresis loops in NiFe/Cu multilayered nanowires,” J. Appl. Phys., vol. 103, Apr. 2008, Art. No. 07D919.  T. Ohgai, X. Hoffer, L. Gravier, J.-E. Wergrowe, and J.-Ph. Ansermet, “Bridging the gap between template synthesis and microelectronics: Spin-valves and multilayers in self-organized anodized aluminum nanopores,” Nanotechnol., vol. 14, pp. 978–982, 2003.  M. Darques, A.-S. Bogaert, F. Elhoussine, S. Michotte, J. de la Torre Medina, A. Encinas, and L. Piraux, “Controlled growth of CoCu nanowires and application to multilayered CoCu/Cu nanowires with selected anisotropy,” J. Phys. D: Appl. Phys., vol. 39, pp. 5025–5032, 2006.  L. Clime, F. Béron, P. Ciureanu, M. Ciureanu, R. W. Cochrane, and A. Yelon, “Characterization of individual ferromagnetic nanowires by in-plane magnetic measurements of arrays,” J. Magn. Magn. Mater., vol. 299, pp. 487–491, 2006.  M. Knobel and K. R. Pirota, “Giant magnetoimpedance: Concepts and recent progress,” J. Magn. Magn. Mater., vol. 242, pp. 33–40, Apr. 2002.  K. Wiesauer and G. Springholz, “Fabrication of semiconductor nanostructures by nanoindentation of photoresist layers using atomic force microscopy,” J. Appl. Phys., vol. 88, pp. 7289–7297, Dec. 2000.  S. Fusil, L. Piraux, S. Mátéfi-Tempfli, M. Mátéfi-Tempfli, S. Michotte, C. K. Saul, L. G. Pereira, K. Bouzehouane, V. Cros, C. Deranlot, and J.-M. George, “Nanolithography based contacting method for electrical measurements on single template synthesized nanowires,” Nanotechnol., vol. 16, pp. 2936–2940, 2005.  S. I. Kiselev, J. C. Sankey, I. N. Krivorotov, N. C. Emley, R. C. Schoelkopf, R. A. Buhrman, and D. C. Ralph, “Microwave oscillations of a nanomagnet driven by a spin-polarized current,” Nature, vol. 425, pp. 380–383, Sep. 2003.  J.-E. Wegrowe, A. Fábián, Ph. Guittienne, X. Hoffer, D. Kelly, J.-Ph. Ansermet, and E. Olive, “Exchange torque and spin transfer between spin polarized current and ferromagnetic layers,” Appl. Phys. Lett., vol. 80, pp. 3775–3777, May 2002.