Experimental demonstration and analysis of random field effects in ferromagnet/antiferromagnet bilayers

Abstract

More than 30 years ago, Malozemoff (Phys. Rev. B 35, 3679 (1987)) hypothesized that exchange interaction at the interface between a ferromagnet (F) and an antiferromagnet (AF) can act as an effective random field, which can profoundly affect the magnetic properties of the system. However, until now this hypothesis has not been directly experimentally tested. We utilize magnetoelectronic measurements to analyze the effective exchange fields at Permalloy/CoO interface. Our results cannot be explained in terms of quasi-uniform effective exchange fields, but are in agreement with the random-field hypothesis of Malozemoff. The presented approach opens a new route for the quantitative analysis of effective exchange fields and anisotropies in magnetic heterostructures for memory, sensing and computing applications.

Exchange bias without directional anisotropy in Permalloy/CoO bilayers

“…These results are consistent with our experimental data. They also explain a surprising recent observation that the effective exchange field in CoO/Py bilayers with thin Py is spatially uncorrelated only for M H cool , but is correlated on the scale of the magnetic exchange length for the opposite direction of M [15]. Indeed, the distribution in Fig.…”

“…The multilayers were deposited in a 150 Oe in-plane magnetic field, which is known to facilitate magnetic ordering in CoO. Py and Ta were deposited by dc sputtering from the stoichiometric targets in 1.8 mTorr of ultrapure Ar, while CoO was deposited from a Co target by reactive sputtering in a mixture of ultrapure oxygen and Ar, with the partial pressure of oxygen optimized as in our previous studies of CoO-based systems [15,18,33,36]. Measurements below the Nèel temperature T N = 290 K of CoO, were performed after cooling from room temperature (RT) in the presence of an external magnetic field H dc = 0.5 kOe.…”

“…In a qualitatively different approach, Malozemoff suggested that frustrated exchange interaction at F/AF interfaces can be described as a random effective exchange field acting on AF [14], which breaks AF up into Imry-Ma domains [29]. Conversely, frustrated F/AF exchange should also manifest as effective random fields acting on F, which has been experimentally confirmed [15,30,31]. For small AF thicknesses, Malozemoff predicted a crossover to the "Heisenberg domain state" (HDS), in which the AF domain sizes become smaller than the domain wall (DW) widths, so the Nèel order becomes twisted everywhere [32].…”

“…Thin-film ferromagnet/antifferomagnet (F/AF) heterostructures may provide an alternative to spin glasses that can combine frustration effects with efficient control and detection of the magnetization state [12]. Indeed, F layers in such heterostructures are amenable to magnetoelectronic readout and can be manipulated by spin torque [13], while the frustration of exchange interaction at the F/AF interface, generally expected due to the incompatibility between F and AF magnetic orderings [14,15], can result in glassy behaviors stabilizing a multitude of magnetization states [13,16,17]. Furthermore, the effects of frustrated exchange interaction in such systems can be tailored by the magnetic properties and thicknesses of individual layers, and by the strength of their coupling [18].…”

Nearly ideal memristive functionality based on viscous magnetization dynamics

“…The studied sample with the structure NiO(8 nm)/Py(10 nm)/Ta(5 nm) was deposited on an oxidized Si substrate by high-vacuum magnetron sputtering at room temperature. NiO was deposited by reactive sputtering from a Ni target in Ar/O 2 mixture, with the partial oxygen pressure adjusted to optimize the magnetic properties, as in our previous studies of NiO and the structurally similar CoO [19][20][21][22] . Py and Ta were deposited in ultrahigh purity Ar.…”