Giant Magnetoresistance in Co-Cu/Cu Multilayers Prepared by Various Electrodeposition Control Modes

Abstract

The giant magnetoresistance ͑GMR͒ effect was studied on electrodeposited Co-Cu/Cu multilayers of 300 bilayer repeats which were produced in an electrochemical cell with homogeneous current distribution from a bath with two solutes (CoSO 4 ,CuSO 4 ). The preparation employed the conventional potentiostatic/potentiostatic and galvanostatic/galvanostatic, as well as an unprecedented galvanostatic/potentiostatic ͑G/P͒ control. We find that the specific deposition parameters rather than the deposition mode itself are decisive for the magnitude of the GMR which could be as high as 10% measured at 1 kOe on substrate-free multilayers in optimized G/P conditions. For this new deposition mode, detailed studies on the dependence of GMR on Co and Cu layer thicknesses as well as the bath pH were performed. No oscillatory behavior of the GMR as a function of the Cu layer thickness could be observed. The results suggest the importance of a Co-dissolution and/or a Co vs. Cu exchange reaction after completing the deposition of each magnetic layer. These reactions lead to the formation of a Cu or Cu-rich interface layer prior to the electrochemical deposition of the actual Cu layer during the subsequent pulse in either deposition mode. It turned out that the properties of this interfacial layer ͑thickness, degree of chemical intermixing͒ strongly influence the resulting GMR behavior of the multilayer.

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

“…First, three ED Co/Cu multilayer sets with various Co and Cu layer thicknesses (while keeping the total multilayer thickness at 300 nm, see Tables I and II) as well as with various total multilayer thicknesses from 50 nm to 300 nm (while keeping the individual layer thicknesses constant at Co(1.1 nm)/Cu(4.0 nm) and Co(2.0 nm)/Cu(4.0 nm), see Table III) were prepared. The fourth set of samples consisted of Co/Cu multilayers with two combinations of fixed layer thicknesses (1.1 nm or 2.0 nm for Co and 4.0 nm for Cu) and a fixed total multilayer thickness of 300 nm whereby the Cu deposition potential was varied from E EC Cu = −600 mV to −250 mV, to a value where a significant dissolution of the magnetic layer 25 is expected to occur (see Table IV). It should be noted that for G/P multilayers deposited at E EC Cu , the actual layer thicknesses correspond fairly well to the nominal values as determined above which was evidenced from detailed XRD studies.…”

“…When measuring GMR on multilayer films with the usual van der Pauw geometry, 21 these effects do not become easily evident but applying a four-point-in-line method for GMR measurements on narrow multilayer strips can clearly reveal them. 22,23 For eliminating these deleterious effects and preparing laterally homogeneous deposits, a tubular cell has been designed 24,25 in which the cathode is at the bottom of the cell with an upward looking deposition area filling the whole cross section of the cell.It is evident from the foregoing discussion that in previous studies the preparation of the ED Co/Cu multilayers from the sulfate bath was not optimal in every respect (and the same holds true also for studies using different bath formulations 4 ). It appeared, therefore, worthwhile to carry out a systematic study of GMR on ED Co/Cu multilayers from the pure sulfate bath under well-controlled conditions which include (i) the use of a smooth Si/Cr/Cu substrate obtained by evaporating nanometer-scale Cr and Cu layers on a Si wafer, (ii) the deposition of a multilayer with a constant total thickness of only 300 nm, (iii) the use of an electrochemical cell ensuring very good lateral homogeneity 24,25 and (iv) the application of an optimized galvanostatic/potentiostatic (G/P) pulse combination.…”

“…22,23 For eliminating these deleterious effects and preparing laterally homogeneous deposits, a tubular cell has been designed 24,25 in which the cathode is at the bottom of the cell with an upward looking deposition area filling the whole cross section of the cell.…”

Giant magnetoresistance in electrodeposited Co-Cu/Cu multilayers: Origin of the absence of oscillatory behavior

“…An aqueous electrolyte containing 0.8 mol/ CoSO 4 , 0.015 mol/ CuSO 4 , 0.2 mol/ H 3 BO 3 and 0.2 mol/ (NH 4 ) 2 SO 4 was used to prepare magnetic/non-magnetic Co/Cu multilayers by using a G/P pulse combination 39 in which a galvanostatic (G) and a potentiostatic (P) pulse is applied for the deposition of the magnetic and the non-magnetic layer, respectively. The Cu deposition potential was optimized according to the method described in Ref.…”

“…The electrodeposition was performed in a tubular cell of 8 mm x 20 mm cross section with an upward looking cathode at the bottom of the cell. 33,39 This arrangement ensured a lateral homogeneity of the deposition current density over the cathode area. Throughout the series, the Cu layer thickness was varied from 0.5 nm to 4.5 nm in steps of about 0.1 nm whereby the magnetic layer thickness was held constant at 2.7 nm (the actual values varied between 2.5 and 3.0 nm).…”