HIGH FREQUENCY MAGNETOIMPEDANCE OF FeNi/Cu/FeNi SENSITIVE ELEMENTS WITH DIFFERENT GEOMETRIES

In this work magnetoimpedance (MI) behaviour was studied experimentally for Fe19Ni81(175 nm)/Cu(350 nm)/Fe19Ni81(175 nm) sensitive elements deposited by rf-sputtering. A constant magnetic field was applied in plane of the sandwiches during deposition perpendicular to the Cu-lead in order to induce a magnetic anisotropy. Sandwiches with different width (w) of FeNi parts were obtained. The complex impedance was measured as a function of the external magnetic field for a frequency range of 1 MHz to 700 MHz for MI elements with different geometries. Some of MI experimental data are comparatively analysed with finite elements numerical calculations data. The obtained results can be useful for optimization of the design of miniaturized MI detectors. Introduction Magnetoimpedance (MI) is the change of the high frequency impedance of a soft ferromagnet under application external magnetic field [1-4]. Mathematical description of the magnetoimpedance phenomenon requires analytical solution of the Maxwellґs equations that can be done only for simplest geometries and using approximations [5-7]. For example, analytical solution is impossible in case of MI sandwiched structure ferromagnet/conductor/ferromagnet for narrower central conductive part. The finite elements method (FEM) was proposed as useful numerical method for complex geometry of MI sensitive elements [8-9]. In this work the longitudinal MI behaviour was studied for Fe19Ni81/Cu/Fe19Ni81 and Fe19Ni81/Fe19Ni81 multilayered sensitive elements of different width in a geometry most appropriate for whole cell type biodetectors. Frequency dependencies of experimental MI ratios measured at reasonably low frequencies, convenient for technological applications are comparatively analysed with FEM numerical calculations data. Experimental The multilayered FeNi/Cu/FeNi and FeNi/FeNi multilayered sensitive elements were deposited onto glass substrates at room temperature by rf-sputtering. The deposition speed was VFe-Ni = 0.38 nm/s for Permalloy (Fe19Ni81) and VCu = 0.22 nm/s for copper. Between the deposition of each layer (including deposition of FeNi/FeNi multilayered elements) a technological 10 min stop was made for surface passivation. MI elements had following dimensions: 1 mm × 8.2 mm with square contacts terminations of 2 mm × 2 mm for copper part; lengths of magnetic parts was kept constant being of 8 mm. Sandwiches with different width (w) of FeNi parts and two layered FeNi/FeNi structures were obtained (see Table and Fig. 1(a)). During multilayered structures deposition a constant magnetic field of about 100 Oe was applied in plane of the samples and perpendicular to the Cu-lead, creating a transverse anisotropy. For example, the anisotropy field Hk ≈ 8 Oe for S1 sample being estimated from the shape M(H) curves obtained by MOKE studies. The complex impedance of the samples (absolute value of the total impedance (Z), real (R) and imaginary (X) components) was measured as a function of the external magnetic field for a frequency (f) range of 1 MHz to 700 MHz for MI elements with different geometries. The sample Solid State Phenomena Online: 2009-04-16 ISSN: 1662-9779, Vols. 152-153, pp 373-376 doi:10.4028/www.scientific.net/SSP.152-153.373 © 2009 Trans Tech Publications, Switzerland This is an open access article under the CC-BY 4.0 license (https://creativecommons.org/licenses/by/4.0/) Table . Description of the multilayered samples Simple MI multilayered structure Width (mm) Type of the structure S1 Fe19Ni81(175 nm)/Cu(350 nm)/Fe19Ni81(175 nm) w1 = 12 Experimental S2 Fe19Ni81(175 nm)/Cu(350 nm)/Fe19Ni81(175 nm) w2 = 9 Experimental S3 Fe19Ni81(175 nm)/Cu(350 nm)/Fe19Ni81(175 nm) w3 = 6 Experimental S4 Fe19Ni81(175 nm)/Cu(350 nm)/Fe19Ni81(175 nm) w4 = 3 Experimental S5 Fe19Ni81(175 nm)/Cu(350 nm)/Fe19Ni81(175 nm) w5 = 1 Model S6 Fe19Ni81(175 nm)/Fe19Ni81(175 nm) w6 = 3 Experimental was connected by conductive silver paint to the microstrip line with 50 Ω of characteristic impedance and the complete test fixture was situated between the two microstrip lines terminated in SMA connectors. The magnetoimpedance was measured using impedance Network Analyzer (Agilent E8358A) by method described in Ref [10]. MI ratios were defined as follow: ∆Z/Z = (Z(H) Z(H = 0))/Z(H = 0), ∆R/R = (R(H) R(H = 0))/R(H = 0) and ∆X/X = (X(H) X(H = 0))/X(H = 0). The sensitivities with respect to applied field was defined as follow: s(∆Z/Z) = d(∆Z/Z)/dH and s(∆R/R) = d(∆R/R)/dH. Results and discussion Fig 1(b). shows the example of field dependence of R and Z components of total impedance for selected frequency of 500 MHz. The same dependences were obtained and analyzed in order to collect the maximum values of ∆Z/Z, ∆R/R and ∆X/X ratios (∆Z/Zmax, ∆R/Rmax and ∆X/Xmax) which for low frequencies appeared in the field closed to Hk. The increase of the frequency resulted in the shift of the maxima toward the higher fields. Decrease of the width w for the same width of the conductive lead results in significant increase of the MI ratio. Rather complex shape of X(H) Figure 1. Schematic description (cross section) of MI multilayered structures used for magnetic, MI measurements and modeling (a); examples of the field dependence of real (main graphs) and imaginary components (inset) of FeNi/Cu/FeNi sensitive elements of different widths. (a)