3b2v7 MAGMA : 8345 Prod:Type: com ED: CM;Brr pp:123ðcol:fig::NILÞ PAGN: ananth SCAN: Sujatha ARTICLE IN PRESS 1 3 Journal of Magnetism and Magnetic Materials ] (]]]]) ]]]­]]] 5 7 Low frequency magnetic noise in epitaxial 9 antiferromagnetically coupled Fe/Cr multilayers 11 V.V. Pryaduna, R. Guerreroa, F.G. Alieva,*, R. Villara, A. Volodinb, C. Van 13 Haesendonckb, I. Vavrac a 15 Dpto. de Fisica de la Materia Condensada, Instituto Ciencia de Materiales ``Nicolas Cabrera'', C-III, Universidad Autonoma de Madrid, 28049 Madrid, Spain b 17 Laboratorium voor Vaste-Stoffysica en Magnetisme, Katholieke Universiteit Leuven, B-3001 Leuven, Belgium c Institute of Electrical Engineering, Slovak Academy of Sciences, 84239 Bratislava, Slovakia 19 21 Abstract 23 For antiferromagnetically coupled epitaxial ½Fe=Crð0 0 1Þ 10 multilayers we detected a strong enhancement of the magnetism-related electrical noise in the vicinity of the orientation transition between the easy and hard axes. Our 25 measurements are performed at different temperatures and we also identified the noise caused by depinning of domain walls (DWs). We are able to detect and follow in real time the motion of rather extended (of the order of 100 mm) 27 DWs by comparing the magnetic noise in the presence and absence of a DC transport current, respectively. The presence of large and small (o1 mm) DWs is confirmed by magnetic force microscopy images obtained at room 29 temperature. r 2001 Published by Elsevier Science B.V. 31 Keywords: Multilayers; Domain wall; Stray field; Magnetic imaging 33 Knowledge of the magnetism-related electrical noise The time (up to 400 s) dependence and the noise 57 35 in antiferromagnetically (AF) coupled magnetic multi- power spectrum (0:01­2 Hz) have been measured for an layers (MML) is important both from an applied and a in-plane magnetic field varying between 600 and 59 37 fundamental point of view. However, until now only one þ600 Oe in steps of 4­10 Oe: Fig. 1 shows the variation study has been reported concerning the low-frequency of the magnetoresistance (part a) as well as of the slope 61 39 noise in Co/Cu MML at room temperature [1]. Here, we A (part b) of the low-frequency part of the noise power present the time dependence and the noise power spectrum (S ¼ A=f ; with f being the frequency) when 63 41 spectrum of the electrical transport in Fe/Cr multilayers the magnetic field is applied along the hard (1 1 0) axis. for a wide temperature interval ranging between 300 and In Fig. 1(b), we observe a strong enhancement of the 65 43 10 K: The epitaxial ½Fe=Crð1 0 0Þ 10 multilayers are magnetic noise around 300 Oe; i.e., within the field prepared in a molecular beam epitaxy (MBE) system region corresponding to the orientation transition (OT) 67 45 on MgO(1 0 0) substrates held at 501C: The Fe layers between the easy axis and the hard axis. A strong have a thickness of 30 (A; while the thickness of the Cr enhancement of the magnetic noise also occurs for fields 69 47 layers, 13:5 (A corresponds to the first AF peak in the below 150 Oe: We link this low-field noise to the interlayer exchange coupling, producing a maximum depinning of domain walls (DWs). Both the depinning 71 49 giant 73 51 75 53 *Correspo UNCORRECTED PROOF magnetoresistance (GMR) which is about 20% at of DWs and the OT are clearly visible in the 300 K and 100% at 4:2 K: A detailed description of magnetoresistance (see Fig. 1(a)). A reproducible ob- sample preparation and structural characterization has servation of these effects can be made at low tempera- been reported elsewhere [2]. tures after the inversion of the magnetic field. The OT is absent when the magnetic field is oriented along the easy axis (Fig. 2) and in that case the electrical transport 77 nding author. Tel.: +341-397-4756; fax: +341- 55 397-3961. noise appears to be dominated by depinning and motion E-mail address: farkhad.aliev@uam.es (F.G. Aliev). of DWs. When the temperature is increased above 79 0304-8853/01/$ - see front matter r 2001 Published by Elsevier Science B.V. P II: S 0 3 0 4 - 8 8 5 3 ( 0 1 ) 0 0 7 4 4 - 2 MAGMA : 8345 ARTICLE IN PRESS 2 V.V. Pryadun et al. / Journal of Magnetism and Magnetic Materials ] (]]]]) ]]]­]]] 1 57 (a) 3 59 4.40 1.0x10-6 5 61 ) 7 ) (V 63 hmO T=10K 5.0x10-7 tage 9 R( 4.35 170 Oe 65 Vol 180 Oe 11 OT 190 Oe 67 0.0 13 -400 -200 0 200 400 69 15 (b) 4.0x10-20 71 -5.0x10-7 T=77K, I=0 17 DWs 73 ) 19 2 V 0 100 200 300 400 75 A( 2.0x10-20 Time (s) 21 77 Fig. 3. Time dependence of the voltage noise caused by electromagnetic induction at 77 K: The magnetic field is 23 79 oriented along the easy axis. The inset shows an MFM image 0.0 obtained at zero magnetic field at room temperature for an area 25 of 60 60mm2: 81 -400 -200 0 200 400 27 H (Oe) 83 100 K; the magnetic field interval where the OT and 29 Fig. 1. (a) Resistance and (b) slope A of the noise power spectra DWs affect the noise diminishes, and instead a non- 85 (see text) for the ½Feð30 (AÞ=Crð13:5 (AÞ 10 magnetic multilayer as reproducible noise signal appears. a function of a magnetic field which is applied along the hard 31 (1 1 0) axis. The measurements are taken at T ¼ 10 K: We have compared the magnetism-related noise signal 87 in the presence and absence of a DC transport current. 33 We find that there exist two qualitatively different 89 contributions to the noise induced by the DWs. When 35 performing the measurements with the field applied 91 4.64 along the easy axis and at temperatures around 100 K 37 (see inset to Fig. 2 which shows slope A at T ¼ 77 K as a 93 T=77K H // 100 function of the magnetic field without applied current), 39 it is possible to discriminate between the noise produced 95 by DWs occurring on different length scales. The DWs 41 created on relatively small length scales (of the order of 97 4.62 1 mm) when compared to the distance between the ) 43 sample voltage probes, move incoherently and therefore 99 hmO only contribute to the current-induced noise. The 45 R( corresponding noise power spectrum is consistent with 101 1E-18 I=0 the spectrum in Fig. 1(b) and scales with the square of ) 47 2 V 1E-20 A( the electrical current. On the other hand, we are also 103 4.60 1E-22 able to detect and follow the real time movement of 49 105 1E-24 0 200 H (Oe) 51 107 -400 -200 0 200 400 53 Fig. 2. UNCORRECTED PROOF rather extended (of the order of 100 mm) DWs. When the noise measurements are done in the absence of a DC current, the motion of the extended DWs causes the appearance of an additional noise voltage due to the H (Oe) electromagnetic induction originating from the stray 109 Resistance of a function of magnetic field directed along field of the DWs. Fig. 3 shows the relative change of the 55 (1 0 0) at T ¼ 77 K. The inset shows slope A at T ¼ 77 K as a induced voltage Vp dF/dt (F is the flux created by 111 function of magnetic field without applied current. the stray field of one or a few extended DWs through the MAGMA : 8345 ARTICLE IN PRESS V.V. Pryadun et al. / Journal of Magnetism and Magnetic Materials ] (]]]]) ]]]­]]] 3 1 loop formed by the probes and the sample surface) as a different dimensions down to the micrometer scale. As function of time after the magnetic field is changed in expected, the domains in the MFM images disappear 23 3 steps of 10 Oe: At T ¼ 77 K; the induction-related noise when applying a magnetic field. The results of our voltage is consistently observed for the narrow magnetic electrical noise measurements indicate that extended 25 5 field range above the characteristic magnetic field which 1801 or 3601 DWs (looking like ``rivers'' in the MFM removes the smaller scale DWs. At low temperatures images) can be held responsible for the induction-related 27 7 (T ¼ 10 K), the strong pinning potential slows down the noise voltage (no DC current), while the smaller scale DWs dynamics and the induced voltage becomes hard to DWs can account for the DW magnetoresistance as well 29 9 detect. At high temperatures (T ¼ 300 K), the DWs as for the current-induced electrical noise. become much more mobile and the appearance of the 31 11 induction-related noise voltage can no longer be linked The authors thank A. Levanuyk for fruitful discus- to a characteristic magnetic field. sions. The work has been supported by Spanish MCYT 33 13 The presence of large and small domain walls in our (BFM2000-0016). ½Fe=Crð1 0 0Þ 10 multilayers is confirmed by the magnetic 35 15 force microscopy (MFM) images obtained at room temperature. The inset in Fig. 2 shows a typical MFM References 37 17 image taken with a commercial MFM system (Park Scientific Instruments, M5) for an area of 60 60 mm2: [1] H.T. Hardner, et al., Phys. Rev. B 48 (1993) 468. 39 19 We observe the irregularly shaped domain walls with [2] R. Schad, et al., Appl. Phys. Lett. 64 (1994) 3500. 41 21 UNCORRECTED PROOF