Ballistic versus diffusive magnetoresistance of a magnetic point contact

Abstract

The quasiclassical theory of a nanosize point contacts (PC) between two ferromagnets is developed. The maximum available magnetoresistance values in PC are calculated for ballistic versus diffusive transport through the area of a contact. In the ballistic regime the magnetoresistance in excess of few hundreds percents is obtained for the iron-group ferromagnets. The necessary conditions for realization of so large magnetoresistance in PC, and the experimental results by García et al are discussed.PACS numbers: 74.80. Dm, 74.50.+r, In recent experiments on study of Ni-Ni and Co-Co point contacts (PC), a surprisingly high negative magnetoresistance exceeding 200% has been discovered. 1,2 The set up of the experiment was typical for observation of giant magnetoresistance (GMR), the effect observed earlier in hybrid systems involving ferromagnetic and normal multilayer metals. 3,4 However, for the multilayer structures the typical change of the resistance reached 10% − 30%, which is considerably lower than the corresponding values of Refs. 1,2. So, one can come easily to the conclusion that the main contribution to the MR comes from the region of the PC.A negative magnetoresistance can be due to scattering on domain walls (DW) and this effect has been considered in a number of works 5-8 giving typical values of MR in a range of few percents. Such considerably low values of the MR were obtained assuming that realistic widths of the DW were large, which resulted in a low scattering amplitudes. Considering sharp DW in the ballistic regime one comes to values ∼ 70%. 8 The fact that a sharp DW may give large MR was used in Ref. 2 to explain the anomalously large values of MR in the experiments on the point contacts. 1,2 However, the theory in Ref. 2 is the perturbation theory, it can not be applied to the explanation of 300% effect. The diminishing of the width of DW, when decreasing the size of the constriction was demonstrated by Bruno. 9 The DW width becomes comparable with PC length, and magnetization rotates almost abruptly inside the constriction. This conclusion holds until the diameter of PC is smaller than its actual length. With further increase of constriction size (diameter) the wall will bend outside of PC, and simple energy considerations show that the DW width will be of the order of PC size. 10

Tunable Conductance of Magnetic Nanowires with Structured Domain Walls

“…In contrast, a series of recent experiments on magnetic nanostructures, and particularly nanowires, revealed that the magnetoresistance in the presence of DWs can be as large as several hundreds [4,5] or even thousands [6,7] of percents. These observations have decisive consequences in so far as DWs are controllable efficiently by applying a magnetic field [8] and can also be steered by a spin-polarized electric current [9], meaning that the magnetoresistance of the structure containing DW is controllable via an electric field.The interpretation of the huge magnetoresistance of DWs observed in magnetic semiconductor (in the ballistic quantum regime) relies on the relative sharpness of DWs on the scale set by the wavelength of carriers (electrons or holes) [10,11,12,13,14]. In such a situation spin-dependent scattering of carriers from DWs is greatly enhanced.…”

“…The interpretation of the huge magnetoresistance of DWs observed in magnetic semiconductor (in the ballistic quantum regime) relies on the relative sharpness of DWs on the scale set by the wavelength of carriers (electrons or holes) [10,11,12,13,14]. In such a situation spin-dependent scattering of carriers from DWs is greatly enhanced.…”

Universal scaling of magnetoconductance in magnetic nanocontacts (invited)

“…If domain wall width is shorter than the mean distance between spin-flip scattering d s ϭv F T s , where v F is the Fermi velocity and T s is the spin relaxation time, the spin can be preserved and large MR can be obtained even in the diffusive transport regime. 9 Furthermore, Tagirov et al have predicted that the nanocontact of highly spin-polarized metals such as LSMO and CrO 2 would show very large MR of 1000% or higher. 10 The mean free path l of CrO 2 is about 1.4 nm or less at room temperature, and this was derived from theoretical values of Lewis et al 21 The value is very small compared to that of normal metals, which is in the range of a few tens of nanometers.…”

“…9 and 10. According to the quasiclassical theory of spin transport through magnetic nanocontact given by Tagirov et al, 9,10 large MR can be obtained if strong spin scattering at the DW is achieved for antiparallel alignment of magnetization in the magnetic nanocontact. To realize large spin scattering, the electron spin orientation should be preserved during transit through nanocontacts.…”

“…Despite differences in formulation, several theories [6][7][8][9][10] suggest that the effect is due to spin scattering at the domain wall and is dependent of the ratio of the spin relaxation length with the domain wall width. Since the domain wall width is a function of the junction area, the effect is consequently a function of the contact area as well.…”