Can a small voltage flip magnetic coupling between thin layers?
Pumping hydrogen into a metal spacer with a small gate voltage reversibly weakens and shifts the coupling between two magnetic layers, even flipping it from parallel to antiparallel.
Source
Voltage control of magnetic order in RKKY coupled multilayers
Study at a glance
- Design
- Other — Sputtered Co/Pt–Ru wedge–Co/Pd multilayers with a GdOx proton-conducting gate; polar MOKE hysteresis loops measured before and after gate-voltage pulses at positions along the Ru thickness wedge.
- N
- Measurements on many gated electrodes along a Ru thickness wedge (39 electrodes for the thickness scans), plus selected devices near RKKY zero crossings; no single sample count.
- Population
- Thin-film magnetic heterostructures with a Ru interlayer 0.4–1.7 nm thick, grown on silicon
- Outcome
- Interlayer exchange field (sign and size of RKKY coupling) versus Ru thickness and gate voltage; field-free switching of the free layer
Structured fields used in claim comparison tables when every cited study has a complete layer.
What they did
The authors grew a stack of two perpendicular magnetic multilayers (a soft Co/Pt layer and a hard Co/Pd layer) separated by a ruthenium spacer whose thickness varies across the sample, topped with a proton-conducting GdOx layer and gold gate. Applying a positive gate voltage splits water from ambient humidity and drives hydrogen into the stack; a negative voltage removes it. They measured the exchange field from minor hysteresis loops using the magneto-optical Kerr effect, across the thickness wedge, before and after gating, and tested short pulses and field-free switching.
What they found
Hydrogen loading reduced the amplitude of the oscillating RKKY coupling and slightly shifted its phase, changing the exchange field by up to 800 Oe. Near the zero crossings of the oscillation, this was enough to switch devices from ferromagnetic to antiferromagnetic coupling or the reverse, and a negative voltage restored the original state over several cycles. A 750 μs voltage pulse was enough to switch coupling, and the voltage alone could flip the free layer by 180° without any applied field.
The limits
What it doesn't show
The mechanism (hydrogen changing the Fermi surface at the interface) is inferred, not directly measured; the authors only exclude a simple thickness-expansion explanation from the direction of the phase shift. Switching efficiency degraded after roughly twenty cycles, which the authors attribute to hydrogen building up in the hard layer, so long-term endurance is unproven. The gated state relaxes back when grounded, so a holding voltage is needed, and the work relies on ambient humidity as the hydrogen source.
Key terms
- RKKY interaction
- An indirect exchange coupling between magnetic layers carried by conduction electrons in a non-magnetic spacer; it oscillates between ferromagnetic and antiferromagnetic as spacer thickness changes.
- Synthetic antiferromagnet
- Two ferromagnetic layers coupled antiparallel through a spacer so their magnetisations largely cancel.
- Magneto-ionics
- Controlling magnetic properties by using a voltage to move ions (here hydrogen) into or out of a magnetic material.
- Exchange field
- The effective field one magnetic layer exerts on another through interlayer coupling, measured here as a shift of the soft layer's minor hysteresis loop.
- Magneto-optical Kerr effect (MOKE)
- A change in the polarisation of reflected light that depends on a sample's magnetisation, used to record hysteresis loops.
Flashcards
0 of 10 answers reviewed
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
What carries the voltage effect into the magnetic stack?
Common questions
Why does the sign change only happen for some devices?
Hydrogen mainly shrinks the oscillation and shifts it slightly, so only spacer thicknesses near a zero crossing of the RKKY oscillation get pushed across zero into the opposite sign.
How do the authors know the phase shift is not just the spacer getting thicker?
Hydrogen expanding the lattice would shift the oscillation in the opposite direction to what was observed, so they attribute it to changes in electronic structure instead.
Why use two layers with different coercivities?
If both layers switched at the same field you could not measure ferromagnetic coupling; making one soft and one hard lets minor loops of the soft layer reveal the exchange field's size and sign.
More on Magnetism and spintronics