Giant Magnetoresistance and Structure of Electrodeposited Co/Cu Multilayers: The Influence of Layer Thicknesses and Cu Deposition Potential

Abstract

The giant magnetoresistance (GMR) and structure was investigated for electrodeposited Co/Cu multilayers prepared by a conventional galvanostatic/potentiostatic pulse combination from a pure sulfate electrolyte with various layer thicknesses, total multilayer thickness and Cu deposition potential. X-ray diffraction (XRD) measurements revealed superlattice satellite reflections for many of the multilayers having sufficiently large thickness (at least 2 nm) of both constituent layers. The bilayer repeats derived from the positions of the visible superlattice reflections were typically 10 -20% higher than the nominal values. The observed GMR was found to be dominated by the multilayer-like ferromagnetic (FM) contribution even for multilayers without visible superlattice satellites. There was always also a modest superparamagnetic (SPM) contribution to the GMR and this term was the largest for multilayers with very thin (0.5 nm) magnetic layers containing apparently a small amount of magnetically decoupled SPM regions. No oscillatory GMR behavior with spacer thickness was observed at any magnetic layer thickness. The saturation of the coercivity as measured by the peak position of the MR(H) curves indicated a complete decoupling of magnetic layers for large spacer thicknesses. The GMR increased with total multilayer thickness which could be ascribed to an increasing SPM contribution to the GMR due to an increasing surface roughness, also indicated by the increasing coercivity. For multilayers with Cu layers deposited at more and more positive potentials, the GMR FM term increased and the GMR SPM term decreased. At the same time, a corresponding reduction of surface roughness measured with atomic force microscopy indicated an improvement of the multilayer structural quality which was, however, not accompanied by an increase of the superlattice reflection intensities. The present results underline that whereas the structural quality as characterized by the surface roughness generally correlates fairly well with the magnitude of the GMR, the microstructural features determining the amplitude of superlattice reflections apparently do not have a direct influence on the GMR. Due to the large giant magnetoresistance (GMR) effect observed in physically deposited Co/Cu multilayers, 1-3 a lot of efforts have been devoted to the study of GMR also on electrodeposited (ED) Co/Cu multilayers (for detailed references, see a recent review).4 A variety of baths have been used for the preparation of ED Co/Cu multilayers, 4 the simplest one containing merely CoSO 4 and CuSO 4 . Over the last two decades, numerous reports have been published on studying the GMR characteristics of ED Co/Cu multilayers from the pure sulfate bath (containing at most some buffering agents). [5][6][7][8][9][10][11][12][13][14][15] In this list of references, we have included only those works from the much larger number of reports 4 in which the ED Co/Cu multilayers were prepared from the sulfate bath at or close to the electrochemically optimized Cu depositi...

Influence of Ag Additive to the Spacer Layer on the Structure and Giant Magnetoresistance of Electrodeposited Co/Cu Multilayers

“…8,61 In spite of the relatively large bilayer thickness ( = 10 nm), these Co/Cu(Ag) multilayers prepared from a perchlorate bath exhibited a GMR effect at least as large as commonly observed in ED Co/Cu multilayers obtained from sulfate-type baths with smaller bilayer repeats. [5][6][7][8][9]53,54,62,63 Thus, the SNMS, XRD and magnetoresistance results are all compatible with the conclusion that the present ED Co/Cu(Ag) multilayers below about 8 at.% Ag incorporated exhibit a multilayer structure even if this layered structure is far from perfect. The imperfections stem mainly from the lateral fluctuations of the bilayer thickness, which increase gradually the interface roughness during the multilayer growth and lead to a corrugation of the layer planes.…”

Influence of Pb Additive to the Spacer Layer on the Structure and Giant Magnetoresistance of Electrodeposited Co/Cu Multilayers

“…The very low intensity of the satellites implies a relatively large error for the peak position determination and this may be one major reason for the deviation of the XRD and nominal bilayer lengths. On the other hand, the ratios XRD / nom ranging from 1.22 to 1.35 correspond well to the general trend by considering previous results on electrodeposited Co/Cu multilayers 37,38 and are unrelated to both the anion being present during the deposition and the element added as surfactant (Pb).…”

Spacer Layer Thickness Dependence of the Giant Magnetoresistance in Electrodeposited Ni-Co/Cu Multilayers

“…It was shown [ 44 , 45 ] that, in the case of Co/Cu multilayers, instead of a targeted Co(3.4 nm)/Cu(2.0 nm) multilayer, the actual thicknesses can be as different as Co(2.0 nm)/Cu(3.4 nm) due to an improperly chosen (not sufficiently negative) Cu deposition potential (in case the Cu layer is deposited under galvanostatic control with definitely lower rate than the diffusion-limited current density, such layer thickness changes always occur [ 28 ]). With the exception of a few cases [ 46 , 47 , 48 , 49 , 50 , 51 , 52 ], the investigation of the spacer layer thickness dependence of the GMR has not been carried out on multilayers prepared by electrochemically-optimized Cu deposition potential. Therefore, the latter reported GMR data do not represent a dependence on the true spacer layer thickness.…”

“…As pointed out in the previous subsection, this requires that the spacer layer should be deposited at the electrochemically-optimized deposition potential and that the GMR FM term should be extracted from the experimental data if there is also a GMR SPM term in the measured magnetoresistance. These conditions were simultaneously fulfilled only for some previous reports on Co/Cu [ 46 , 47 , 48 , 49 , 50 ], Co-Fe [ 51 ] and Ni-Co/Cu [ 52 ]. All of these studies demonstrate the absence of an oscillatory GMR as a function of the spacer layer thickness, whereas there are numerous reports on the observation of an oscillatory GMR in various electrodeposited multilayers (for detailed references, see, e.g., Ref.…”