Electric Field Control of Interfacial Ferromagnetism in ${\mathrm{CaMnO}}_3/{\mathrm{CaRuO}}_3$ Heterostructures

New mechanisms for achieving direct electric field control of ferromagnetism are highly desirable in the development of functional magnetic interfaces. To that end, we have probed the electric field dependence of the emergent ferromagnetic layer at ${\mathrm{CaRuO}}_3/{\mathrm{CaMnO}}_3$ interfaces in bilayers fabricated on ${\mathrm{SrTiO}}_3$. Using polarized neutron reflectometry, we are able to detect the ferromagnetic signal arising from a single atomic monolayer of ${\mathrm{CaMnO}}_3$, manifested as a spin asymmetry in the reflectivity. We find that the application of an electric field of $600\mathrm{kV}/\mathrm{m}$ across the bilayer induces a significant increase in this spin asymmetry. Modeling of the reflectivity suggests that this increase corresponds to a transition from canted antiferromagnetism to full ferromagnetic alignment of the ${\mathrm{Mn}}^{4+}$ ions at the interface. This increase from $1{\mu }_B$ to $2.5–3.0{\mu }_B$ per Mn is indicative of a strong magnetoelectric coupling effect, and such direct electric field control of the magnetization at an interface has significant potential for spintronic applications.

Figure 1

(Top) Heterostructure schematic with plan view AFM of a typical bilayer after deposition of the ${\mathrm{CaMnO}}_3$ layer. (Bottom) Magnetic hysteresis loop of a $[({\mathrm{CaRuO}}_3{)}_3/({\mathrm{CaMnO}}_3{)}_9{]}_{10}$ superlattice at 10 K. (Inset) Magnetization vs temperature of the same superlattice in an applied field of 20 mT after field cooling in 5 T.

Figure 2

(a) Spin dependent reflectivity of sample A after cooling to 10 K in 700 mT under a $-400\mathrm{V}$ bias. The solid lines represent a model fit to the data. (b) Spin asymmetry of sample A after cooling in 700 mT without a bias. (c) Spin asymmetry of sample A after cooling in 700 mT with a bias of $-400\mathrm{V}$. Solid lines in (b) and (c) are spin asymmetries predicted by modeling. (d) Nuclear and magnetic depth profile used to model sample A at $-400\mathrm{V}$. (e) Close-up of the highlighted regions of (b) and (c). (f) Spin asymmetry of sample A after cooling in $-400\mathrm{V}$ and 700 mT and varying bias voltage at constant temperature. The measured region corresponds to the FWHM of the spin asymmetry peak shown in part (e). (Inset) Summation of the spin asymmetry of the peak in (f) as a function of applied voltage.

Figure 3

Spin asymmetry of sample B as a function of $Q_Z$ under a voltage of (a) 0 V and (b) $-300\mathrm{V}$. Sample was field cooled to 20 K in a field of 700 mT. The black line in (b) is a guide to the eye based on a simple toy model with a magnetization of $2.54{\mu }_B/\mathrm{Mn}$ in a single unit cell of ${\mathrm{CaMnO}}_3$ at the interface. (c) Specular reflection rocking curves after cooling in 0 V and $-300\mathrm{V}$. (inset) Specular reflection rocking curves at a constant temperature of 10 K for voltages of $-400$, 0, and $+350\mathrm{V}$. (d) Spin asymmetry of an identically measured ${\mathrm{SrTiO}}_3$ substrate at $-300\mathrm{V}$ and $+300\mathrm{V}$.