banner image
No data available.
Please log in to see this content.
You have no subscription access to this content.
No metrics data to plot.
The attempt to load metrics for this article has failed.
The attempt to plot a graph for these metrics has failed.
Inductive determination of the optimum tunnel barrier thickness in magnetic tunneling junction stacks for spin torque memory applications
Rent this article for
View: Figures


Image of FIG. 1.
FIG. 1.

(Color online) (a) Schematic picture of the MgO based MTJ stack. refers to the exchange coupling between the free layer (FL) and the reference layer (RL), refers to the antiferromagnetic interlayer exchange coupling between the reference layer (RL) and the pinned layer (PL) whereas refers to the exchange bias coupling between the antiferromagnetic pinning layer (AF) and the pinned layer (PL) (b) TMR and RA product as a function of MgO barrier thickness measured by CIPT technique (Ref. 9). Line shows the exponential thickness dependence of RA.

Image of FIG. 2.
FIG. 2.

(Color online) (a)–(c) PIMM data (open dots) for t MgO = 0.76 nm at different easy axis fields. Lines show fits by an exponentially damped sinusoid. (d) Static field dependence of the precession frequency derived from PIMM (open dots). Line shows the dispersion relation of a Stoner-Wolfarth single-domain model with HK  = 1.8 mT and JFL  = 20.1 μJ/m2. Inset: MOKE loop (black dots) and single domain model approximation (line).

Image of FIG. 3.
FIG. 3.

(Color online) MgO thickness dependence of (a) the exchange coupling JFL between the free layer and the reference layer (circles) as well as the uniaxial anisotropy energy KFL of the free layer (squares) and (b) the effective damping parameter. Open (full) symbols in (a) are referred to values of JFL or KFL derived from PIMM (MOKE) measurements. Line shows the Néel coupling behavior derived by fitting JFL data to Eq. (3). In (b) open triangles refer to the damping parameter derived at parallel configurations (α P), open squares are referred to the damping parameter at antiparallel configurations (α AP), and full squares are referred to the averaged damping parameter between both configurations. At MgO thickness t MgO > 0.76 nm both configurations have the same effective damping parameter (α AP = α P  ≡ α).

Image of FIG. 4.
FIG. 4.

(Color online) (a), (e) Dispersion relation and minor MOKE loops; (b), (f) effective damping dependence on easy axis magnetic fields; and (c–d), (g–h) simulated magnetic field dependence of magnetization orientation of each ferromagnetic layer for (c), (f) easy axis (e.a.) and (d), (h) hard axis (h.a.) magnetic fields with t MgO = 0.88 nm [(a), (b), (c), and (d)] and t MgO = 0.71 nm [(e), (f), (g) and (h)]. (a), (e) Open dots (line) show the measured (simulated) resonance frequency. Inset: Full dots (line) shows the measured (simulated) minor MOKE loops. (c–d), (g–h) FL, RL, and PL refer to the free layer, reference layer, and pinned layer respectively.

Image of FIG. 5.
FIG. 5.

Schematic picture of the cross-sectional profile of the MTJ stack. The Néel coupling induces a FM coupling between the reference layer and the magnetic moments located at the “valleys” of the rough free layer. For thin enough MgO barriers the Néel coupling is so strong that those magnetic moments (small black arrows) tend to be aligned along the reference layer magnetization (white arrow). For pararallel configurations (a) the whole free layer is parallel aligned to the reference layer, whereas for antiparallel configurations (b) just a fraction of the free layer is reversed and a region with a large inhomogeneous distribution of the magnetization is developed (dashed area).


Article metrics loading...


Full text loading...

This is a required field
Please enter a valid email address
752b84549af89a08dbdd7fdb8b9568b5 journal.articlezxybnytfddd
Scitation: Inductive determination of the optimum tunnel barrier thickness in magnetic tunneling junction stacks for spin torque memory applications