Improving the efficiency of current-induced motion of chiral domain walls by interfacial engineering

Abstract

The present invention relates to a magnetic domain wall motion memory cell (racetrack memory device) that includes a 4d or 5d metal dusting layer (DL) at the ferromagnetic / heavy metal interface of a ferromagnetic (FM) structure or a synthetic antiferromagnetic (SAF) structure of the basic racetrack device structure.

Description

[Technical Field]

[0001] The present invention relates to memory storage systems, and in particular to memory storage systems that use electric current to move magnetic domain walls within a magnetic racetrack. [Background technology]

[0002] Racetrack memory devices have attracted interest as high density storage devices. These devices are disclosed, for example, in U.S. Patent No. 6,275,999. More advanced racetrack memory cells have already been developed, including vertical nanowire storage elements, as disclosed, for example, in U.S. Patent No. 6,275,999.

[0003] In particular, racetrack devices based on chiral domain wall (DW) magnetic bits in perpendicularly magnetized ferromagnet / heavy metal thin film systems are promising candidates for next-generation spintronic memories (Patent Documents 3 and 4). These currents can be used to manipulate magnetic bits encoded in magnetic spin textures (domains, skyrmions, or antiskyrmions) within nanoscale racetracks.

[0004] Current-induced domain wall motion (CIDWM) has evolved significantly from in-plane magnetic to synthetic antiferromagnetic (SAF) racetracks due to advances in large-capacity spin-transfer torque (STT) and spin-orbit torque (SOT) mechanisms. Driven by chiral spin torque arising from spin-orbit coupling in the presence of inversion symmetry breaking at the ferromagnet / heavy metal (HM) interface, Néel domain walls in thin films with strong perpendicular magnetic anisotropy (PMA), stabilized by the Dzyaloshinskii-Moriya interaction (DMI) at the ferromagnet / HM interface, can be moved along the current direction at high speeds in both straight and curved racetracks (see, for example, Patent Document 5). The high-speed and energy-efficient motion of such magnetic bits along 2D or 3D racetracks driven by spin currents remains a key challenge for their commercial implementation.

[0005] Even more efficient DW motion has been reported in a synthetic antiferromagnetic (SAF) racetrack, consisting of two perpendicularly magnetized ferromagnetic subracetracks antiferromagnetically coupled across an ultrathin ruthenium layer. The giant exchange coupling torque (ECT) in the SAF structure provides a further dominant driving mechanism, enabling increased DW propagation velocities exceeding approximately 1,000 m / s. ECT in rare-earth transition metal alloys is further maximized at the angular momentum compensation temperature of ferrimagnetic alloys. Recently, efficient CIDWM has also been found in certain magnetic insulators.

[0006] For example, by varying the HM layer in contact with the interfacial ferromagnetic layer or by tuning the thickness of the ferromagnetic layer, significant progress has been made in terms of a detailed understanding of the influence of magnetic chirality on the interface-derived chiral spin torque and the mechanisms underlying CIDWM. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] US-B6,834,005 [Patent Document 2] US-A2014 / 204648 [Patent Document 3] US-A2014 / 0009994 [Patent Document 4] US-A2014 / 0009993 [Patent Document 5] EP3171364A1 Summary of the Invention [Problem to be solved by the invention]

[0008] However, for commercially viable high-speed and low-power racetrack memory devices, the domain wall velocity is still too low and the threshold current density is still too high. Therefore, it was an object of the present invention to provide more efficient current-induced domain wall motion. [Means for solving the problem]

[0009] The present invention significantly reduces the threshold current density and greatly increases the efficiency of domain wall motion by introducing an atomically thin 4d or 5d, preferably 4d metal "dusting" layer (DL) at the ferromagnetic / heavy metal (HM) interface. In a further preferred embodiment, a subatomic-thick dusting layer of Pd and Rh, more preferably just one monolayer thick, is introduced at the HM / ferromagnetic interface, thereby increasing the domain wall velocity by up to 3.5 times at a given current density. The Néel DW moves more than three times faster at the same injection current density compared to an otherwise identical structure without the DL. Furthermore, the threshold current density J, defined as the minimum current density required to overcome the effective pinning field and move the DW, is significantly reduced. th is substantially reduced by incorporating atomically thin DLs.

[0010] Without wishing to be bound by this theory, we believe that such improvements are due to the fine interrelationships between tailored spin-orbitronics parameters, i.e., parameters resulting from spin-orbit coupling effects; specifically, the Dzyaloshinskii-Moriya interaction and uniaxial magnetic anisotropy. This work demonstrates how important interfacial modification is to enable tailored racetracks with enhanced chiral domain wall motion efficiency, and it directly demonstrates the close internal correlation between the Dzyaloshinskii-Moriya interaction and uniaxial magnetic anisotropy. [Brief explanation of the drawings]

[0011] [Figure 1-1]Engineered FM and SAF racetrack memory structures with interfacial dusting layers. Figure 1a shows a schematic of the FM (upper panel) and SAF (lower panel) racetrack structures with a dusting layer (DL) inserted between the heavy metal (Pt) and ferromagnetic metal (Co) layers. Figure 1b shows the upper panel: a schematic of the atomic stack of the FM / DL / HM structure along the fcc(111) direction; the lower panel: the color-coded elements used in the FM / DL / HM stack; orange, green, and light blue correspond to FM, DL, and HM, respectively. The numbers in the upper right corner of each element's rectangle correspond (from top to bottom) to the spin-orbit coupling constant, spin diffusion length, and Stoner criterion parameter, respectively, and the numbers in the lower left corner are the corresponding in-plane lattice constants in fcc(111) units. Figure 1c shows scanning electron microscopy of a typical racetrack. Figure 1d shows a cross-sectional HRTEM image. Figure 1 e Corresponding EDX mapping of the FM film with a 0.1 nm thick Pd dusting layer, highlighting the presence of the atomically thin Pd layer. [Figure 1-2] Continued from Figure 1-1. [Figure 2-1] Interfacial DL-engineered chiral domain wall motion in FM and SAF structures. Figure 2a-f. Current-induced DW motion in FM (left panel) and SAF (right panel) structures with various DL materials: Pd (a and b, orange background), Rh (c and d, violet background), and Ir (e and f, blue background). The insets in (a) and (b) illustrate typical Kerr images of domain wall motion in response to a series of injected current pulses (approximately 1.0 × 10 A / cm), consisting of twelve (a) and four (b) 10 ns pulses, respectively, in FM and SAF samples with a 0.1 nm Pd DL, ensuring that all DWs move in the direction of current injection. The difference in image contrast arises from the difference in magnetization in the FM and SAF samples, with bright and dark areas representing down (

number

number

number

number

[0012] The basic structure of the racetrack device according to the invention is based on a ferromagnetic (FM) structure or a synthetic antiferromagnetic (SAF) structure.

[0013] The ferromagnetic structure comprises one or more, preferably two or three, layers of ferromagnetic material. When the ferromagnetic structure comprises more than two layers, preferably no two adjacent layers are identical. The ferromagnetic layers are: Fe, Co, Ni, or Mn, or - alloys of Fe and / or Co, or an alloy of Ni which may further contain one or more of Fe and Co, or an alloy of Mn, which may optionally further contain one or more of Fe, Co or Ni, or - a compound comprising Fe and at least one of Ir, Al, Si, or Ge, and - Compounds containing Mn and Si The magnetic material is made of a ferromagnetic material selected from one or more of:

[0014] Typically, the ferromagnetic structure may have a total thickness in the range of 0.5 to 1.5 nm, preferably 0.75 to 1.3 nm, more preferably 0.9 to 1.2 nm. Each individual layer of the ferromagnetic structure may independently have a thickness in the range of 0.1 to 1.5 nm, preferably 0.12 to 1.0 nm, more preferably 0.14 to 0.8 nm.

[0015] The ferromagnetic layer of the ferromagnetic structure is preferably sandwiched between the HM layer and the coupling layer.

[0016] The HM layer comprises at least one of Pt, Ir, W, Ta, or Ru. The HM layer may advantageously have a thickness of 0.8 nm to 2.0 nm, preferably 1.0 nm to 1.8 nm, more preferably 1.2 nm to 1.7 nm.

[0017] The bonding layer comprises at least one of Ru, W, Ta, or Ir. The bonding layer may advantageously have a thickness of 0.4 nm to 1.5 nm, preferably 0.6 nm to 1.0 nm, more preferably 0.7 nm to 0.9 nm.

[0018] The composite antiferromagnetic structure can consist of two ferromagnetic layers antiferromagnetically coupled through a coupling layer. In a preferred embodiment, the FM / coupling layer / FM sandwich is deposited on an HM layer. The ferromagnetic layer, coupling layer, and HM layer are as described above. Preferably, the coupling layer is made of Ru or Ir.

[0019] According to the present invention, the FM and / or SAF racetrack memory structure includes an interfacial dusting layer. This dusting layer is made of a 4d or 5d metal, preferably a 4d metal, and most preferably a metal with a long spin diffusion length, preferably greater than 5 nm. It is further preferred that the metal exhibits an fcc structure. Most preferred are Pd and Rh, while Ir and Ru are less preferred.

[0020] The dusting layer is preferably located at the HM / FM interface, ie, the dusting layer is preferably sandwiched between the HM and FM layers.

[0021] The thickness of the dusting layer is preferably in the range of 0.1 to 1.5 nm, more preferably 0.1 to 1.0 nm, and most preferably 0.2 to 0.7 nm. This means that the dusting layer can have dimensions in the range of subatomic layer thicknesses, defined as being within or even lower than the lattice constant of the selected DL material. In some embodiments, the dusting layer has a thickness in the range of about the diameter of an atom of the DL material, i.e., about 0.2 nm; this thickness is referred to as "one monolayer." "Subatomic layer" thicknesses and "one monolayer" thicknesses are equivalent thicknesses predicted from sputter deposition rates.

[0022] The individual layers of the racetrack device, including the dusting layer, can be deposited, for example, by magnetron sputtering, preferably at room temperature, onto silicon wafers, which are preferably thermally oxidized so that they are covered with a SiO2 layer (about 30 nm). The layers are preferably sandwiched between a bottom TaN layer (about 2 nm) and a capping TaN layer (about 5 nm), both of which are highly resistive. The deposition parameters of these materials can be calibrated, for example, by quartz crystal microbalance and X-ray reflectometry. Layer thicknesses are expressed in square units (e.g., nm 2The sputtering rate can be determined and controlled by the amount of material (gram atoms) sputtered per gram. Racetrack nanowires can be fabricated using photolithography and argon ion milling. Domain walls (DWs) can be created within the racetrack nanowires by injecting pulses of current in the presence of an external longitudinal magnetic field.

[0023] Applicable The present invention has applications in many different fields of technology, such as spintronics, including, but not limited to, magnetic random access memories; magnetic recording hard disk drives; magnetic logic devices; security cards using magnetically stored information; semiconductor devices in which large magnetic fields provided by domain wall fringing fields can be used to locally change the electronic properties of semiconductors or semiconductor heterostructures; and mesoscopic devices in which the electron energy levels therein are sufficiently small that they can be substantially affected by the application of local magnetic fields.

[0024] Advantages of the Invention In the model of an ideal racetrack with a homogeneous film stack, there is no threshold current for CIDWM. However, in reality, the threshold current J th must be applied to depin the DW by thermally assisted excitation across an energy barrier that prevents the DW from moving freely within the wire. When a specific DL is introduced, the uniaxial anisotropy energy K u and coercive field H C This barrier is reduced, as shown by the reduction of J th However, when the DL becomes thicker, a decrease in θ SH The inevitable decline in th In the case of SAF, J th It is important to notice that the FM cases are sufficiently similar to those in the corresponding FM cases, which means that J th This suggests a key role for the LM layer in determining

[0025] To enable the application of CIDWM-based racetrack memory devices, low J th Both high DW speeds are often required.

number

number

number

number

[0026] In conclusion, the present invention provides a novel method for substantially reducing the current required to both depin and move chiral domain walls within magnetic racetracks. The method involves the insertion of atomically thin 4d or 5d element dusting layers, preferably made from 4d elements with an fcc structure, at critical interfaces, preferably FM / HM interfaces, within magnetic multilayers that form simple ferromagnetic or synthetic antiferromagnetic racetracks. A clear linear correlation between the perpendicular magnetic anisotropy exhibited by the ferromagnetic racetracks and the Dzyaloshinskii-Moriya interaction that drives the chirality of the domain walls was found for both 4d and 5d element insertion layers. These findings are realized through controlled manipulation of interfacial spin-orbit coupling.

[0027] The invention is illustrated by the following examples. [Example]

[0028] Atomically thin dusting layers Two sets of structures were prepared by DC magnetron sputtering at room temperature as shown in Figure 1a: ferromagnetic structures consisting of Co(0.3 nm) / Ni(0.7 nm) / Co(0.15 nm) sandwiched between a Pt(1.5 nm) HM layer and a Ru(0.85 nm) coupling layer, hereafter referred to as FM structures; composite antiferromagnetic structures consisting of a Co(0.3 nm) / Ni(0.7 nm) / Co(0.15 nm) lower ferromagnetic layer and a Co(0.5 nm) / Ni(0.7 nm) / Co(0.15 nm) upper ferromagnetic layer deposited on the same Pt(1.5 nm) HM layer and antiferromagnetically exchange coupled through a Ru(0.85 nm) coupling layer, hereafter referred to as SAF structures. A series of atomically thin layers of Pd, Ir, Rh, and Ru (hereafter referred to as dusting layers DL) with thicknesses ranging from 0.1 to 0.7 nm were inserted directly onto the Pt HM layer in both structures before the deposition of the ferromagnetic material. A schematic diagram of the FM and SAF structures is shown in Figure 1a, with the elemental dusting layers illustrated in Figure 1b. CIDWM was studied in a typical racetrack, 3 μm wide and 50 μm long, fabricated by photolithography and Ar ion milling (Figure 1c). The motion of individual DWs within these nanowires in response to voltage pulses of fixed length (approximately 10 ns) was detected using Kerr microscopy. The DW positions within the nanowires before and after the pulse injection were recorded and used to determine the DW velocity along the racetrack.

[0029] A cross-sectional high-resolution transmission electron microscope (HRTEM) image of the FM structure with a 0.1 nm palladium dusting layer is shown in Figure 1d. This image reveals a highly (111)-oriented structure of the face-centered cubic (fcc) thin film structure along the out-of-plane direction. The extremely smooth surface of the film is confirmed by atomic force microscopy. A high-angle annular dark-field scanning TEM (HAADF-STEM) image and associated energy-dispersive X-ray spectroscopy (EDX) map of the layered structure in Figure 1e directly reveal the Pd dusting layer at the predicted location between the Pt and Co / Ni / Co layers, even though the inserted Pd DL is only 0.1 nm thick.

[0030] Current-induced chiral DW motion The chiral spin torque drives the DW along the direction of the injected current, regardless of the DL parameters. Distinctly different CIDWM behaviors are observed in FM and SAF structures, which are sensitively determined by the DL material and thickness. For the Pd DL, the threshold current density J required to observe DW motion is th is found to be significantly reduced in FM structures with Pd DLs as thin as 0.1 nm (Figure 2a), while DW velocity increases across the entire range of current densities examined. The maximum current density that can be applied to the racetrack is limited by the formation of multiple magnetic domains, likely due to increased nanowire temperature, as previously observed in comparable resistive nanowires. As the Pd layer thickness increases, the maximum current density decreases, along with a degradation of CIDWM performance. In contrast, for SAF structures, Figure 2b shows that the introduction of a Pd DL improves CIDWM efficiency at all DL thicknesses examined, resulting in significantly higher DW velocities at otherwise identical current densities compared to SAF structures without a DL (reference SAF). The efficiency of DW motion is maximized for Pd DL thicknesses of approximately 0.2 nm, with DW velocities up to approximately 1000 m / s, as shown in Figure 2b. This velocity is up to 3.5 times higher than the same structure without a Pd DL at the same current density.

[0031] The racetrack with the Rh dusting layer behaves similarly to the Pd DL case. The DL thickness range for which CIDWM is enhanced is up to 0.4 nm for FM and substantially reduced J thThe DW velocity is extended with increasing current density (Figure 2c). The PMA for the 0.7 nm Rh DL was too weak to allow CIDWM measurements. For the SAF shown in Figure 2d, CIDWM is enhanced for 0.1 and 0.2 nm thick Rh DLs. However, for thicker Rh layers (>0.3 nm), the DW velocity actually decreases as the current density increases. While the PMA was too weak for the FM case at 0.7 nm to successfully detect DW motion, a stable single DW still forms in the SAF case as a result of exchange coupling, but no significant CIDWM could be observed before thermal nucleation of random DWs occurs.

[0032] A significantly different behavior is observed when an Ir DL is used. In the case of FM, J th decreases when the Ir layer is 0.1 nm thick, but then increases almost linearly with further increases in Ir thickness. At 0.1 nm Ir, only a very slight enhancement in CIDWM rate was observed at low current densities, and CIDWM was otherwise slower than in the reference FM sample. For the SAF structure, a systematic degradation of CIDWM occurs as soon as the Ir DL is inserted (see Figure 2f). However, J th It is worth mentioning that decreases smoothly for Ir layers with thickness up to 0.5 nm, but increases dramatically for thicker layers.

[0033] To directly compare the effect of different dusting layers on the performance of CIDWM, the threshold current density J th and DW velocity for both FM and SAF structures with DL thickness t in Figures 3a-3d. DL As a function of 8 A / cm 2 The dramatic role of DLs is easily seen, most notably for Pd and Rh DLs. In these cases, the CIDWM efficiency increases with increasing current density at small t DLHowever, with the Ir dusting layer, the CIDWM efficiency increases only slightly for 0.1 nm and then decreases for t DL As increases, it decreases monotonically to zero propagation (Figures 3a and 3b).

[0034] To distinguish CIDWM from DW creep, which occurs even at small current densities, a threshold current density J th was defined as the current density at which the chiral DW velocity exceeds 5 m / s. th A decrease in J was observed for the racetracks with DLs, except for the Ir case. It is worth mentioning that the spacing between adjacent fcc(111) planes in the dusting layer is about 0.22 nm, which corresponds to the center of the colored region in Figure 3. A plausible argument is that the efficiency of the chiral DW motion is maximized by inserting a one-monolayer-thick 4d metal (Pd and Rh) DL. Specifically, the most promising case is the SAF structure with a 0.2 nm-thick Pd DL layer: J compared to the reference sample. th A substantial decrease of about 70% and an increase of about 350% in CIDWM velocity indicate a very fast DW velocity of up to about 1000 m / s, resulting in a 1.2 × 10 12 A / m 2 This was observed at a typical current density of .

[0035] Magnetic field dependence of DW velocity CIDWM originates from the chiral spin torque, where the chirality of the DW in both FM and SAF structures is stabilized by the interfacial DMI arising from the HM layer with strong spin-orbit coupling. Hence, the DW velocity is sensitively determined by the magnetic field applied along the racetrack: the external longitudinal magnetic field H xis added to or subtracted from the DMI effective field that stabilizes the chiral DW. The DW velocity was measured as a function of the longitudinal magnetic field. For clarity, the motion of the DW with ↓↑ and ↑↓ domain configurations under positive current is shown in Figure 4, where racetracks incorporating Pd, Rh, and Ir DLs with thicknesses of 0.1, 0.2, 0.3, and 0.5 nm are presented. The data are for a fixed current density (1.2 × 10 8 A / cm 2 ) at zero external field, both the ↓↑ and ↑↓ DWs move in the same direction as the injected current for both the FM and SAF structures, which is consistent with SOT and ECT. For the FM structures presented in Figures 4a-d, the DW velocity is H x , which is typical behavior as previously observed for this reference structure. The slope depends on the DL material and thickness, and the sign reverses for ↓↑ and ↑↓ DW, i.e., v(↓↑,H x ,J)=v(↓↑,-H x ,J). This behavior can be easily understood within a 1-D DW analytical model. Such behavior has two main features: vH x The slope of the curve and the DW velocity tending towards zero H x Therefore, vH x By fitting the curves, magnetic properties that cannot be measured directly can be predicted, in particular the interfacial DMI strength D.

[0036] In the case of SAF, v(↓↑,H x ,J)=v(↓↑,-H x ,J), with the symmetry effect of DW chirality on the DW velocity, we obtain a completely different profile of vH xA curve is observed. The coupled Néel DW in the SAF structure undergoes a more complex response in the presence of an external field. Several magnetic properties contribute to the detailed shape of the response curve. A 1-D model is used to reproduce the experimental results and extract these parameters. Figures 4b, 4d, 4f, and 4h show that the variation of the DW velocity with the external field increases with the thickness of the inserted Pd and Rh DLs. However, the DW velocity is less sensitive to the thickness of the Ir DL, which may be correlated with the relatively weak spin Hall effect in Ir. Notably, as the thickness of the Ir DL increases beyond 0.5 nm (Figure 4h), the corresponding vH x The curves show the vH for thinner Ir DLs. x The mirrored profile of the curves is shown in Fig. 1. We attribute this opposite dependence of the DW velocity to the reported opposite sign of the interfacial DMI at the Co / Ir interface compared to the Co / Pt interface.

[0037] Magnetic properties To understand the dependence of CIDWM on the dusting layer material and thickness, the saturation magnetization M s , the effective perpendicular magnetic anisotropy constant

number

number

number

number

[0038] t DL M to r / M s The dependency of M with DL insertion is shown in Fig. 5b. r / M s The change in ratio is seen in Fig. 5a for FM, as can be seen in Fig. 5b for the M s It has been previously shown that the CIDWM in SAF samples is largely due to the large ECT that increases, and more similar are the two sublayer moments, namely, M r / M s = 0. The reference SAF sample without DL is M r / M s Note that,is optimized to be close to zero.,As discussed above, the efficiency of CIDWM depends on the,specific DL, which is M,. r / M s Even if we deviate from zero, it increases. r / M s When is adjusted to zero by small modifications to the racetrack structure, even faster CIDWM in the DL-engineered racetrack is expected.

[0039] The fastest DW velocity v in the SAF structure lgis achieved with the help of an external longitudinal field as shown in Figure 4. lg The DL thickness dependence of v is plotted in Fig. 5d. For the Pd and Rh dusting layers, v lg increases rapidly and then saturates with increasing DL thickness. However, for Ir DL, v lg remains approximately constant with a gradual decrease as the DL thickness increases. These results indicate that the DL used here is x , suggesting that they can be divided into two groups with respect to their response to the DMI field.

[0040] In the FM structure, a monotonic decrease in the DMI constant D is observed with increasing DL thickness, except for a slight increase in the case of Rh at about 0.1 nm (Fig. 5e). It is worth mentioning that there is an obvious sign change in D with increasing Ir DL thickness.

[0041] Manufacturing and measurement methods Sample preparation and DW velocity measurements The samples were deposited by magnetron sputtering at room temperature on a Si wafer coated with a SiO2 (thermal silicon oxide) layer (approximately 30 nm). All of these samples were sandwiched between a bottom TaN layer (approximately 2 nm) and a capping TaN layer (approximately 5 nm), both of which have high resistivity. The deposition parameters of these materials were calibrated using a quartz crystal microbalance and X-ray reflectometry. 50 μm x 3 μm racetrack nanowires were fabricated using photolithography and argon ion milling. All injected current pulses were fixed at a duration of approximately 10 ns. The DW velocity was determined from Kerr microscopy measurements. DWs can be created within the racetrack nanowires by injecting pulses of current in the presence of an external longitudinal magnetic field.

[0042] TEM sample preparation and investigation Cross-sectional TEM specimens were prepared by conventional preparation methods. First, the cross sections were mechanically polished on both sides. They were then thinned to an electron watermark by dimple grinding on one side and polishing with Ar ions at 5 kV on the other side in a Gatan PIPS (Precision Ion Polishing System) system (Gatan, Pleasanton, USA). For HR-TEM / STEM investigations, an FEI TITAN80-300 electron microscope (FEI, Hillsboro, USA) with a probe collector was used at an accelerating voltage of 300 kV. EDX experiments were performed using a Super-X detector system (four silicon drift detectors (Oxford, Abingdon, UK) symmetrically positioned around the sample area inside the objective lens), which was installed on the microscope for faster X-ray speeds and better collection efficiency. The acquired EDX maps were analyzed and processed using Bruker Esprit software (Bruker, Billerica, USA).

[0043] Magnetic property measurement and calculation of the DMI constant D The magnetization of the samples was measured in a superconducting quantum interference device (SQUID) at room temperature.

number

number

number

[0044] v from a 1-D analytical model under steady-state motion conditions lg Derivation of From a 1-D analytical model based on DW moments that describes the DWM of the SAF system, the DW velocity

number

number

number

number

[0045] In the formula, α i is the attenuation parameter of each sublayer, where i corresponds to the L (lower) or U (upper) layer; M i is the magnetization; β i is the non-adiabatic constant; u i is the STT-related (spin transfer torque) DW velocity; γ is the gyromagnetic ratio; Δ is the DW width;

number

number

number

[0046] When an external longitudinal field is applied to the system, as shown in the first equation, the velocity increases as the SOT increases due to the lower layer experiencing a larger SOT than the upper layer. L = 0 or π. A simple assumption is that M L =M U =M, and if we ignore the STT-related terms, then ψ L By taking the condition =0 or π, the above set of equations can be rewritten into the following form:

number

number

number

[0047] In the case of Pd and Rh, due to the long spin diffusion length in these materials, the SOT generated from the Pt sublayer is not significantly attenuated as it passes through the DL, and therefore v lg is directly proportional to the DW width, and therefore can be expected to show a similar saturation behavior of DW width with increasing thickness. In the case of Ir, the SOT decreases greatly with increasing DL thickness, with only a slight modification of the DW width. Therefore, v lg has small variations with varying DL thickness.

Claims

1. A magnetic domain wall motion memory cell (racetrack memory device) comprising a 4d metal dusting layer (DL) at the ferromagnetic / heavy metal interface of a ferromagnetic (FM) structure or a synthetic antiferromagnetic (SAF) structure of a basic racetrack device structure; The ferromagnetic (FM) structure is at least one layer of ferromagnetic (FM) material; a heavy metal (HM) layer, and bonding layer Including, The synthetic antiferromagnetic (SAF) structure comprises two layers of ferromagnetic (FM) material that are antiferromagnetically coupled via a coupling layer, thus forming an FM / coupling layer / FM sandwich that is deposited on a HM layer; the dusting layer comprises a 4d metal; The memory cell.

2. The memory cell of claim 1 , wherein the dusting layer has a thickness in the range of 0.1 to 1.5 nm.

3. The memory cell of claim 1 or 2, wherein the 4d metal exhibits an fcc structure.

4. The memory cell of any one of claims 1 to 3, wherein the 4d metal is Pd or Rh.

5. Ferromagnetic (FM) materials are Fe, Co, Ni, or Mn, or an alloy of Fe and / or Co, or an alloy of Ni containing at least one of Fe and Co; or an alloy of Mn containing at least one of Fe, Co, or Ni; or a compound containing Fe and at least one of Ir, Al, Si, or Ge; and 10. The memory cell of claim 1, wherein the metal is selected from the group consisting of: a compound containing Mn and Si.

6. The memory cell of claim 1 or 5, wherein at least one layer of ferromagnetic material is sandwiched between the HM layer and the coupling layer.

7. The memory cell of claim 1 , 5 or 6 , wherein the HM layer comprises at least one of Pt, Ir, W, Ta, or Ru.

8. The memory cell of any one of claims 1 and 5-7, wherein the HM layer has a thickness of 0.8 nm to 2.0 nm.

9. The memory cell of any one of claims 1 and 5-8, wherein the bonding layer comprises at least one of Ru, W, Ta, or Ir.

10. The memory cell of any one of claims 1 and 5 to 9, wherein the coupling layer has a thickness of 0.4 nm to 1.5 nm.

11. A memory cell according to any one of claims 1 and 5 to 9, characterized in that when the ferromagnetic structure comprises two or more layers, two adjacent layers are either identical or not identical.

12. A magnetic random access memory; a magnetic recording hard disk drive; a magnetic logic device; a security card using magnetically stored information; a semiconductor device; or a mesoscopic device, comprising a memory cell according to claims 1 to 11.

Images