Spin waves imaged inside a nano-oscillator launch harder from one edge than the other

Until now, what happens inside a single spin Hall nano-oscillator had never been watched directly, because the dynamics run on nanometer length scales and inside fractions of a nanosecond ^[6]. The device itself is simple to describe. A direct current through a constriction a few tens of nanometers wide drives the local magnetization into steady precession. The DC input comes out as a tunable radio-frequency signal ^[4].

The movies show the auto-oscillation sitting at both edges of that constriction, with a pronounced asymmetry favouring one edge ^[9], and the emitted spin waves leaving perpendicular to the applied magnetic field ^[10]. Micromagnetic simulations reproduced this only after three effects usually left out were added ^[11]. One is the grain structure of the film, whose boundaries affect the waves ^[12]. Another is a perpendicular anisotropy slightly weakened by nanofabrication; it moves the spots the waves launch from ^[13]. The third is the Dzyaloshinskii-Moriya interaction, an interface effect that makes waves travelling in opposite directions behave differently, and that is what accounts for the one-sidedness ^[14].

Adding three ingredients until a simulation matches a picture is weaker evidence than one prediction tested once, since each ingredient is another free parameter. The interface term is on firmer footing than that, because it predicts a specific directional asymmetry between counter-propagating waves, and a directional asymmetry is what the X-ray images contain ^[14][9]. The phys.org account of the work treats the three-ingredient agreement as a sign that the physical models used to design SHNOs and their synchronized networks need refining ^[15]. How far that shifts any predicted frequency or coupling strength is left open.

"Only by seeing the spin waves directly could we understand which ingredients really govern these oscillators. Effects like grain boundaries and the Dzyaloshinskii-Moriya interaction turned out to be essential, not optional," said Sebastian Wintz, who led the measurements at BESSY II ^[18]. Johan Akerman, who leads the Gothenburg group, put the network claim carefully: "We know that SHNOs can synchronize in very large networks. What this direct imaging gives us is a much firmer microscopic basis for understanding and controlling how spin waves set the coupling and phase relationships within those networks," he said ^[19]. The imaging behind that statement was done on single oscillators ^[6].

The measurement was stroboscopic, using X-ray magnetic circular dichroism for contrast, at a few tens of nanometers of spatial resolution and a time resolution well below the oscillation period near 6 GHz ^[8]. One cycle at 6 GHz lasts about 167 picoseconds ^[20]. It also needs a storage ring: MAXYMUS sits at BESSY II in Berlin ^[2].

Then there is the side result. Under prolonged illumination with intense soft X-rays, the samples' magnetic properties changed gradually and permanently ^[16]. That was reported for the particular stack studied here, an ultrathin CoFeB layer capped with MgO, and not for magnetic materials in general. But the same interface is a standard ingredient of spintronic devices now considered for satellites and other spacecraft. On that basis the report argues that the response of such devices to large radiation doses deserves a closer look ^[17].