Double-current magnetic moment overturning method and device for spin-orbit torque device and spin-orbit torque device

Abstract

The invention provides a dual-current magnetic moment overturning method and device for a spin-orbit torque device and the spin-orbit torque device, and the method comprises the steps: inputting a first current pulse into a spin-orbit torque channel layer in a first direction at a first time, and inputting a second current pulse in a second direction at a second time; wherein a plane formed by intersection of the first direction and the second direction is perpendicular to the vertical magnetic moment; the first current pulse and the second current pulse are overlapped in time sequence and different in end time; the current density of the first current pulse and the second current pulse is determined based on the critical flip current density; the first current pulse and the second current pulse are used for enabling the magnetization orientation of the magnetic moment in the free layer to be subjected to vertical-in-plane conversion, and the magnetization orientation of the converted magnetic moment is certainly converted into a positive polarity direction or a negative polarity direction. According to the invention, field-free deterministic flipping of the vertical magnetic moment can be realized based on an existing SOT device structure, and a double-current injection basis can be provided for application of magnetic storage and Boolean operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of spin electronics, and in particular to a double-current magnetic moment flipping method and device for a spin-orbit torque device and the spin-orbit torque device. BACKGROUND

[0002] Spin-Orbit Torque (SOT) magnetic devices are becoming one of the most competitive non-volatile memory devices in the post-Moore era. The deterministic flipping of the magnetic moment of SOT devices has a wide application prospect in magnetic storage and Boolean logic operations. However, the existing SOT devices mostly need an external magnetic field to assist in the deterministic flipping of the magnetic moment, which brings great challenges to the application and promotion of such devices in the MRAM field. At present, there have been some researches on the deterministic flipping of the magnetic moment of SOT devices under a field-free condition (field-free in this application refers to no external magnetic field) by using double currents: The existing technical document 1 can realize the field-free flipping of the magnetic moment by using the spin-orbit torque in a SOT device with in-plane biaxial anisotropy. The device preparation and magnetic moment flipping process of this technical solution are as follows: (1) depositing a thin film with a structure of Ta(3) / Pt(3) / MgO(2) / PtMn(10) / Pt(5) / Ru(1); (2) processing the deposited thin film into a cross-shaped device by a lithography process; (3) preparing electrodes and contact pads by a lithography and stripping process; (4) after the sample preparation is completed, applying a 1.2 T in-plane magnetic field parallel to the write channel direction, and performing two-hour annealing treatment at a temperature of 300°C to prepare a SOT device; (5) as shown in (a) of Figure 1 , 10 pulses can be applied to the prepared SOT device along the horizontal arm (A→B) to realize the deterministic flipping of the in-plane Néel vector along the direction under a field-free condition; as shown in (b) of Figure 1 , 10 pulses can also be applied to the SOT device along the vertical arm (C→D) to realize the deterministic flipping of the in-plane Néel vector along the direction under a field-free condition. In addition, Figure 1 (c) of shows the change of the anomalous Hall resistance of the SOT device after applying the pulses .

[0003] The scheme in Prior Art Literature 1 can reversibly flip the antiferromagnetic Néel vector in an antiferromagnetic thin film by using the spin-orbit torque, but this technical solution has the following defects: (1) The material used for preparing the thin film in this scheme is an in-plane magnetic anisotropy material, which is not conducive to high-density integration due to the in-plane magnetic anisotropy, and the practical application and promotion of the prepared device has great challenges; (2) The thin film structure prepared by this technical solution is complex and difficult to produce and process; (3) The compatibility of the heterostructure (i.e. the thin film) prepared by this technical solution with the Ta (or W) / CoFeB / MgO / CoFeB tunneling structure which is currently mature in processing technology and has superior performance is difficult to evaluate, and the application prospect of this device on the SOT device with perpendicular magnetic anisotropy is not clear; (4) The magnetic thin film flipped by this device is an antiferromagnetic thin film, not a ferromagnetic thin film required for the free layer in the SOT device.

[0004] Considering that the construction of the current SOT device mainly depends on the magnetic tunnel junction unit with perpendicular magnetic anisotropy, using a device with perpendicular magnetic anisotropy can ensure lower switching current, higher thermal stability and smaller element packaging volume, which is conducive to the further development of high-integration-density memory. For SOT devices with perpendicular magnetic moment, the typical scheme for realizing magnetic moment flipping under field-free conditions by double current is to use the non-traditional magnetic layer in Prior Art Literature 2 and the US patent with the publication number US9741414B2 and the title “Spin orbit and spin transfer torque-based spintronics devices”.

[0005] Prior Art Literature 2 can realize the field-free flipping of the perpendicular magnetic anisotropy magnetic moment by controlling the angle between the current vector and the three-fold symmetry axis based on the CuPt / CoPt heterojunction with three-fold symmetry (three-fold symmetry is unique to the crystal structure itself). The device preparation and magnetic moment flipping scheme of this technical solution is: (1) CuPt and CoPt are grown in sequence on a SrTiO3(111) substrate by magnetron sputtering technology; (2) the structure obtained by growth is processed into a columnar shape by using the combination process of ultraviolet maskless lithography machine and ion beam etching technology; (3) after depositing the electrode by thermal evaporation method, the optical microscope image of the prepared device can be shown as (a) in Figure 2 ; (4) input two currents and at the same time (two currents and can also be input at the same time, or two currents and can also be input at the same time, or only one current is input). As shown in (b) in Figure 2 , two currents and Taking current as an example, the black arrow represents the actual current. and The direction of injection, the red arrow indicates the direction of the current vector sum. Since the CuPt / CoPt heterojunction has triple symmetry, if the direction of the current vector sum is not along the symmetry axis of the crystal structure, the flip symmetry will be broken, and field-free deterministic flipping will be achieved. However, the technical solution in the prior art document 2 has the following defects: (1) The solution involves the introduction of a non-traditional magnetic layer (i.e., CuPt), and it is still challenging whether a magnetic tunnel junction can be further grown on the basis of this heterojunction. (2) The device fabrication process is complex and costly: Since the solution requires the growth of CuPt / CoPt heterojunction with a specific crystal orientation on a single crystal substrate, and requires precise control of sputtering conditions to obtain an ideal triple symmetry structure, it places extremely high demands on process stability and repeatability, which brings huge challenges to the actual promotion and use of SOT devices. Therefore, the application prospects of the SOT device in the prior art document 2 in general applications are still unclear.

[0006] The SOT device structure used in US Patent Publication No. US9741414B2 is "heavy metal layer / free layer / tunnel barrier layer / free layer / heavy metal layer", and as Figure 3 As shown, the operation steps for deterministic magnetic moment reversal using this device are as follows: (1) Current is supplied to the upper heavy metal layer to apply SOT to the upper free layer, thereby changing the direction of the magnetic moment of the upper free layer; (2) Current is supplied to the lower heavy metal layer to apply SOT to the lower free layer, thereby changing the direction of the magnetic moment of the lower free layer. The principle of this technical solution to achieve fieldless deterministic reversal is as follows: when current is supplied to both heavy metal layers at the same time, both the upper and lower free layers will generate a damping-like field. and similar fields These two equivalent fields, each free layer produces and It will assist another free layer and Under such conditions, the required external auxiliary magnetic field may be reduced or even canceled, thereby achieving the simultaneous flipping of the magnetic moments of the two free layers. However, this technical solution has the following drawbacks: (1) The SOT device used in this solution requires two free layers, and the two free layers need to flip simultaneously when the device is working. Whether a magnetic tunnel junction can be further grown under this structure is still a challenge. (2) This solution requires a period of time during which two currents are simultaneously applied, and theoretically, the current application needs to be stopped simultaneously.

[0007] While the second and third schemes can achieve deterministic reversal of magnetic moments in SOT devices under field-free conditions through dual current injection, the devices they employ differ from current SOT structures with perpendicular magnetic anisotropy. Therefore, directly applying these current injection methods to existing SOT device structures presents significant challenges. How to inject dual currents to achieve field-free deterministic reversal of perpendicular magnetic moments on the basis of existing SOT-MRAM structures, thereby expanding the applications of SOT devices in magnetic storage and Boolean logic operations, is a problem that urgently needs to be solved.

[0008] References: Existing technical literature 1: DuttaGupta, S., Kurenkov, A., Tretiakov, OA, et al. Spin-orbit torque switching of an antiferromagnetic metallic heterostructure, Nat. Commun. 11, 5715 (2020). Existing technical document 2: Zhao, T., Zheng, Z., Wang, J., et al. Spin logic enabled by current vector adder, Nat. Commun. 16, 2988 (2025). Summary of the Invention

[0009] In view of this, embodiments of the present invention provide a dual-current magnetic moment reversal method, apparatus and spin orbital moment device for spin orbital moment devices, which can realize field-free deterministic reversal of vertical magnetic moment based on the existing SOT device structure.

[0010] One aspect of the present invention provides a dual-current magnetic moment reversal method for a spin-orbit moment device, the spin-orbit moment device having perpendicular magnetic anisotropy, comprising a magnetic tunnel junction containing a free layer and a spin-orbit moment channel layer stacked below the free layer; the method includes the following steps: A first current pulse is input into the spin-orbit-moment channel layer along a first direction at a first time, and a second current pulse is input into the spin-orbit-moment channel layer along a second direction at a second time, wherein the first direction and the second direction intersect, and the plane defined by the first direction and the second direction is perpendicular to the stacking direction of the magnetic tunnel junction and the spin-orbit-moment channel layer. Within a dual-current injection time step, the first time is earlier than or equal to the second time, the timing of the first current pulse and the second current pulse overlaps, and the end times of the first current pulse and the second current pulse are different; the vector sum of the current densities of the first current pulse and the second current pulse is greater than the critical reversal current density, and the current density of either the first current pulse or the second current pulse is less than the critical reversal current density; furthermore, the first current pulse and the second current pulse are used to cause a vertical-in-plane reversal of the magnetization orientation of the magnetic moments in the free layer, and to deterministically convert the magnetization orientation of the transformed magnetic moments to a positive or negative polarity direction, so that a deterministic reversal of the magnetic moments in the free layer can be achieved without an external magnetic field.

[0011] In some embodiments of the present invention, the vector sum of the current densities of the first current pulse and the second current pulse is greater than the critical switching current density, and the current density of either the first current pulse or the second current pulse is less than the critical switching current density, including: Under the first set condition, the vector sum of the current densities of the first current pulse and the second current pulse is greater than the critical reversal current density, and the current density of the second current pulse is less than the critical reversal current density. Under the second set condition, the vector sum of the current densities of the first current pulse and the second current pulse is greater than the critical reversal current density, and the current density of the first current pulse is less than the critical reversal current density. Under the third setting condition, the current density of the first current pulse is greater than the critical reversal current density, and the current density of the second current pulse is less than the critical reversal current density.

[0012] In some embodiments of the present invention, for the writing function of the spin orbital moment device, the first setting condition includes, within a time step, a first time equal to a second time and the end time of the second current pulse being later than the end time of the first current pulse. For the writing function of the spin orbital moment device, the second setting condition includes, within a time step, a first time equal to a second time and the end time of the second current pulse earlier than the end time of the first current pulse, or, a first time earlier than the second time and the end time of the second current pulse earlier than the end time of the first current pulse. For the writing function of the spin orbital moment device, the third setting condition includes that within a time step, the first time is earlier than the second time and the end time of the second current pulse is later than the end time of the first current pulse. For the Boolean logic function of the spin orbital moment device, the first setting condition includes that within a time step, the first time is equal to the second time, the end time of the second current pulse is later than the end time of the first current pulse, and the logic state corresponding to the first current pulse is the first Boolean logic state. For the Boolean logic function of the spin orbital moment device, the second setting condition includes, within one time step, the first time is equal to the second time, the end time of the second current pulse is earlier than the end time of the first current pulse and the logic state corresponding to the first current pulse is the first Boolean logic state, or, the first time is earlier than the second time, the end time of the second current pulse is earlier than the end time of the first current pulse and the logic state corresponding to the first current pulse is the first Boolean logic state. For the Boolean logic function of the spin orbital moment device, the third setting condition includes that within a time step, the first time is earlier than the second time, the end time of the second current pulse is later than the end time of the first current pulse, and the logic state corresponding to the first current pulse is the first Boolean logic state.

[0013] In some embodiments of the present invention, for the Boolean logic function of the spin orbital moment device, when the logic state corresponding to the first current pulse is the second Boolean logic state, the current densities of the first current pulse and the second current pulse are set based on a set current strategy; wherein, the set current strategy includes: If the first current pulse and the second current pulse overlap at multiple time points on the time axis within a time step, the vector sum of the current density of the first current pulse and the second current pulse is less than the critical flip current density. If the first current pulse and the second current pulse overlap by only one time point on the time axis within a time step, then the current density of the first current pulse and the current density of the second current pulse are both less than the critical flip current density.

[0014] In some embodiments of the present invention, the input direction of the first current pulse corresponding to the first Boolean logic state is on a straight line with the input direction of the first current pulse corresponding to the second Boolean logic state, and the input direction of the second current pulse corresponding to the first Boolean logic state is opposite to the input direction of the second current pulse corresponding to the second Boolean logic state.

[0015] In some embodiments of the present invention, the first current pulse and the second current pulse respectively perform the following functions: If the first current pulse and the second current pulse overlap by only one time point on the time axis, the first current pulse is used to cause a vertical-in-plane reversal of the magnetization orientation of the magnetic moment in the free layer, and the second current pulse is used to definitively change the magnetization orientation of the magnetic moment in the free layer to a positive or negative polarity. Otherwise, the first current pulse and the second current pulse are used together to cause a vertical-in-plane reversal of the magnetization orientation of the magnetic moment in the free layer. If the end time of the second current pulse is later than the end time of the first current pulse, the second current pulse is used to definitively change the magnetization orientation of the magnetic moment in the free layer to a positive or negative polarity. If the end time of the second current pulse is earlier than the end time of the first current pulse, the first current pulse is used to definitively change the magnetization orientation of the magnetic moment in the free layer to a positive or negative polarity.

[0016] In some embodiments of the present invention, the spin orbital moment device further includes a first current axis layer and a second current axis layer respectively connected to the spin orbital moment channel layer, the first current axis layer and the second current axis layer being located above the substrate layer; wherein, the first current axis layer is used to inject a first current pulse into the spin orbital moment channel layer, and the second current axis layer is used to inject a second current pulse into the spin orbital moment channel layer.

[0017] In some embodiments of the present invention, the top view of the spin orbital moment device is cross-shaped or star-shaped.

[0018] Another aspect of the present invention provides a spin-orbit moment device having perpendicular magnetic anisotropy. The device includes: a top electrode, a magnetic tunnel junction located below the top electrode and including a free layer, a spin-orbit moment channel layer located below the free layer, a base layer located below the spin-orbit moment channel layer, and a first current axis layer and a second current axis layer respectively connected to the spin-orbit moment channel layer; the first current axis layer and the second current axis layer are located above the base layer, the first current axis layer is used to inject a first current pulse into the spin-orbit moment channel layer, and the second current axis layer is used to inject a second current pulse into the spin-orbit moment channel layer; The spin orbital moment device is used to achieve deterministic reversal of the magnetic moment within the free layer by the method described in any of the above embodiments.

[0019] Another aspect of the present invention provides a dual-current magnetic moment reversal device for a spin orbital moment device, characterized in that the device includes a driver and a means for generating current pulses, the driver being used to drive the means for generating current pulses to perform the steps of the method described in any of the above embodiments.

[0020] The present invention proposes a dual-current magnetic moment reversal method, device, and spin orbital moment device for spin orbital moment devices. It can use first and second current pulses with different timings to make the magnetic moment in the free layer deviate from the vertical easy axis direction and approach the in-plane direction. Furthermore, it can reverse the magnetization direction of the magnetic moment approaching the in-plane direction to the positive polarity direction (+z) or the negative polarity direction (-z), thereby achieving deterministic reversal of the vertical magnetic moment based on the existing SOT film structure.

[0021] Additional advantages, objects, and features of the invention will be set forth in part in the description which follows, and will also become apparent in part to those skilled in the art upon studying the description, or may be learned by practice of the invention. The objects and other advantages of the invention can be realized and obtained by means of the structures specifically pointed out in the description and drawings.

[0022] Those skilled in the art will understand that the objectives and advantages achievable with the present invention are not limited to those specifically described above, and that the above and other objectives achievable with the present invention will become clearer from the following detailed description. Attached Figure Description

[0023] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, are not intended to limit the scope of the invention. The components in the drawings are not drawn to scale but are merely illustrative of the principles of the invention. For ease of illustration and description of certain parts of the invention, corresponding portions in the drawings may be enlarged, i.e., may appear larger relative to other components in an exemplary device actually manufactured according to the invention. In the drawings: Figure 1 This is a schematic diagram of current injection into an in-plane anisotropic SOT device according to an embodiment of the present invention.

[0024] Figure 2 This is a schematic diagram of dual current injection of a SOT device with triple symmetry in one embodiment of the present invention.

[0025] Figure 3 This is a schematic diagram of dual current injection of a SOT device in one embodiment of the present invention.

[0026] Figure 4 This is a schematic diagram of the structure of a spin orbital moment device in one embodiment of the present invention.

[0027] Figure 5 This is a top view of the spin orbital moment device in one embodiment of the present invention.

[0028] Figure 6 This is a timing diagram of the first current pulse and the second current pulse in one embodiment of the present invention.

[0029] Figure 7This is a simulation diagram of the pulse width of the first current pulse and the second current pulse in one embodiment of the present invention.

[0030] Figure 8 This is a magnetization dynamics trajectory diagram during dual current injection in one embodiment of the present invention.

[0031] Figure 9 This is a schematic diagram of a dual-current-passed spin orbital moment device according to an embodiment of the present invention.

[0032] Figure 10 This is a schematic diagram illustrating the magnetic moment reversal principle during dual current injection in one embodiment of the present invention.

[0033] Figure 11 This is a schematic diagram of dual current injection applied to Boolean logic operations in one embodiment of the present invention. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the embodiments and accompanying drawings. Here, the illustrative embodiments and descriptions of this invention are used to explain the invention, but are not intended to limit the invention.

[0035] It should also be noted that, in order to avoid obscuring the invention with unnecessary details, only the structures and / or processing steps closely related to the solution according to the invention are shown in the accompanying drawings, while other details that are not closely related to the invention are omitted.

[0036] It should be emphasized that the term "including / comprises" as used herein refers to the presence of a feature, element, step, or component, but does not exclude the presence or addition of one or more other features, elements, steps, or components.

[0037] It should also be noted that, unless otherwise specified, the term "connection" in this article can refer not only to a direct connection, but also to an indirect connection involving an intermediary.

[0038] In the following description, embodiments of the invention will be illustrated with reference to the accompanying drawings. In the drawings, the same reference numerals represent the same or similar parts, or the same or similar steps.

[0039] Since the spin-polarized current induced by charge flow can only drive the magnetization direction in the plane, existing SOT devices require an external auxiliary magnetic field parallel to the injected current direction to achieve deterministic magnetic moment reversal. The reversal direction is determined by the relative direction of the external magnetic field and the applied current. Given the unclear application prospects of existing dual-current injection methods for SOT devices in general scenarios such as MRAM, this application proposes a series of dual-current field-free injection methods to achieve directional magnetic moment reversal without an external magnetic field (directional reversal is equivalent to deterministic reversal). By inputting two current pulses with different timings into the SOT device, the synergistic effect between a field-like SOT and a damped SOT generated by the dual currents can be used to directionally reverse the vertical magnetic moment. This method is applicable to SOT devices with vertical magnetic anisotropy and expands the application of SOT devices in magnetic storage and Boolean logic operations. Therefore, the method proposed in this application has potential value for large-scale production and widespread use.

[0040] The spin orbital moment device to which the dual current injection method proposed in this application is applicable has perpendicular magnetic anisotropy, and the device structure can be the same as that of existing SOT devices, such as... Figure 4 As shown, the SOT device applied in this application may include: a substrate layer, a spin-orbit moment channel layer (the spin-orbit moment channel layer is used to generate spin-orbit moments, and in this application, the spin-orbit moment channel layer may be simply referred to as the SOT channel layer) located above the substrate layer, a magnetic tunnel junction located above the SOT channel layer, and a top electrode located above the magnetic tunnel junction. The magnetic tunnel junction includes a free layer, a tunnel barrier layer, and a reference layer. The free layer is located above the spin-orbit moment channel layer, and the reference layer is located below the top electrode. The reference layer and the free layer have perpendicular magnetic anisotropy, and the reference layer has a fixed magnetization direction that cannot be changed by injected current.

[0041] The spin-orbit moment device mentioned in this application can be fabricated using existing materials for fabricating SOT devices, and this application does not specifically limit the materials used for fabricating SOT devices. For example, the substrate layer can be fabricated using materials such as metal (single metal or metal alloy), silicon oxide, or silicon; the SOT channel layer requires a strong spin-orbit coupling effect, and the SOT channel layer can be fabricated using materials capable of generating spin-orbit moments (e.g., strong spin-orbit coupling materials), such as a heavy metal layer fabricated using one or more of tantalum (Ta), platinum (Pt), tungsten (W), and iridium (Ir); the top electrode can serve as the lead for the magnetic tunnel junction and the substrate layer, so the material used for fabricating the top electrode can be one or more combinations of materials such as titanium (Ti), gold (Au), aluminum (Al), copper (Cu), Ta, Pt, tantalum nitride (TaN), and titanium nitride (TiN); moreover, there are no specific limitations on the materials used for fabricating the magnetic tunnel junction with perpendicular magnetic anisotropy used in this application, for example, the tunnel junction can be a widely used Ta (or W) / CoFeB / MgO / CoFeB system. In other words, the SOT device of this application is not limited by the fabrication materials and has broad application prospects.

[0042] As an example, the reference layer in a magnetic tunnel junction can be a combination of magnetic and metallic materials, the tunnel barrier layer can be made of insulating materials such as magnesium oxide and aluminum oxide, and the free layer is a magnetic material layer. Its preparation materials include, but are not limited to, one or more of Fe, Co, FeB, CoB, CoFe, and CoFeB. For example, common alloy ratios (Co:Fe:B) can be 1:8:1, 2:6:2, or 4:4:2. When the free layer is combined with the spin-orbit moment channel layer, it can exhibit perpendicular magnetic anisotropy at room temperature.

[0043] In some embodiments of the present invention, the SOT device involved in this application can be fabricated as follows: using magnetron sputtering technology, a spin orbital channel layer, a free layer, a tunnel barrier layer, a reference layer, and a top electrode layer are sequentially deposited on a substrate layer. Before growing the top electrode, each deposited layer needs to be patterned. Furthermore, the fabrication process may include one or more annealing steps to induce crystallization of the magnetic layers (such as the free layer) and the tunnel barrier layer; the annealing temperature can be selected within the range of 100°C to 400°C. The magnetron sputtering method is merely an example; other physical vapor deposition methods such as molecular beam epitaxy can also be used.

[0044] In some embodiments of the present invention, to clarify the dual current injection channel, this application may further add a current axis layer to the existing spin-orbit moment device with vertical magnetic anisotropy. In this case, the SOT device also includes at least two current axis layers electrically connected to the spin-orbit moment channel layer. These two current axis layers are respectively used to inject two current pulses (i.e., a first current pulse and a second current pulse) with different timings into the spin-orbit moment channel layer. For example, the spin-orbit moment device may also include a first current axis layer and a second current axis layer respectively connected to the spin-orbit moment channel layer, and the first current axis layer and the second current axis layer are located above the substrate layer. The first current axis layer is used to inject a first current pulse into the spin-orbit moment channel layer, and the second current axis layer is used to inject a second current pulse into the spin-orbit moment channel layer.

[0045] The terms "first current axis layer" and "second current axis layer" mentioned in this application only refer to different current axis layers and do not indicate sequence or other meanings. Furthermore, this application uses two current pulses with different timing sequences: "first current pulse" and "second current pulse." The first current axis layer can inject a first current pulse (or a second current pulse) into the spin-orbit moment channel layer, and the second current axis layer can inject a second current pulse (or a first current pulse) into the spin-orbit moment channel layer. For simplicity, the following description uses the example of the first current pulse being input to the spin-orbit moment device through the first current axis layer and the second current pulse being input to the spin-orbit moment device through the second current axis layer. Additionally, if the spin-orbit moment device includes three or more current axis layers, only two current axis layers need to be selected during current injection to input the first and second current pulses respectively, leaving the other current axis layers idle. Figure 5 As shown, when the top view of the spin orbital moment device containing two current axis layers is cross-shaped or star-shaped, the first current pulse can be input along the direction of the solid arrow and the second current pulse can be input along the direction of the dashed arrow.

[0046] As an example, the two current axis layers can be fabricated using the same material as the spin-orbit-moment channel layer (e.g., the current axis layers can be fabricated using the same metallic material used to fabricate the spin-orbit-moment channel layer), or they can be fabricated using a different conductive material than the spin-orbit-moment channel layer. That is, the first and second current axis layers can be fabricated using conductive materials. Furthermore, the first and / or second current axis layers can be integral with the spin-orbit-moment channel layer, and they can also be electrically connected to the spin-orbit-moment channel layer via welding or conductive adhesive. Additionally, this application does not specifically limit the shape and arrangement of the first and second current axis layers, nor does it specifically limit the shape of the SOT device. For example, the first and second current axis layers can be in the form of wires or thin films. The positions of the first and second current axis layers should ensure that the input directions of the first and second current pulses are not parallel (if the first and second directions are parallel, it is the same as the case of single-axis current flow, in which case the torque symmetry cannot be broken, and an external auxiliary magnetic field is required). Furthermore, the plane formed by the first and second current pulses should be perpendicular to the stacking direction of the free layer and the spin-orbit moment channel layer. As another example, the SOT device in this application can be an asymmetrically shaped device.

[0047] In some embodiments of the present invention, the deterministic reversal of the magnetic moment within the free layer can be achieved by using the aforementioned spin orbital moment device to execute the dual current injection method proposed in this application. When a first current pulse and a second current pulse with different injection timings are applied to the spin orbital moment channel layer, a spin current is generated. This spin current, transmitted to the free layer, produces a damped spin orbital moment and a field-like spin orbital moment (SOTs are classified as damped spin orbital moments and field-like spin orbital moments). The synergistic effect of these two types of torques enables the field-free directional reversal of the vertical magnetic moment in the magnetic free layer, allowing the writing of data "0" and "1". Furthermore, the parallelism or antiparallelism of the magnetic moment directions between the free layer and the reference layer affects the tunneling magnetoresistance of the entire device. Data reading can be performed based on the resistance value of the tunneling magnetoresistance (determined by the read current under a certain voltage). The sign of the current pulse (i.e., the direction of the current pulse input to the spin orbital moment channel layer) determines the magnetic moment reversal direction of the magnetic tunnel junction in the SOT device, thereby achieving the control of the magnetization state of the free layer.

[0048] More specifically, the magnetization orientation of the magnetic moment within the free layer is transformed through the damped SOT and field-like SOT generated in the free layer by the first and second current pulses. That is, the precession of the magnetic moment generated by the applied current pulse is due to the damped SOT (i.e., the damped spin orbital moment, which can be expressed as...). ) and field-like SOT (i.e., field-like spin orbital moment, which can be expressed as () is the dominant force. and The following conditions must be met: ; ; in, It is the magnetic moment vector of the free layer. It is the direction vector of the spin-polarized current caused by spin-orbit coupling.

[0049] The dynamic process of the spin orbital moment acting on the magnetic moment can be represented by the LLGS (Landau-Lifshitz-Gilbert-Slonczewsk) equations: ; ; in, The thickness of the ferromagnetic layer, This is the proportionality constant between the damped spin orbital moment and the field-like spin orbital moment. It is the gyrometry ratio. The permeability of free space, This is the equivalent field of perpendicular magnetic anisotropy. Let be the damping constant. The saturation magnetization is For electric charge and current, For the spin Hall angle, To reduce Planck's constant, This represents the fundamental charge.

[0050] In some embodiments of the present invention, the magnetic moment can be directionally flipped by setting the first and second current pulses according to specific conditions. Furthermore, the vertical magnetic moment can be flipped towards either the positive or negative polarity direction by adjusting the input direction of the current pulses. The dual-current injection method proposed in this application specifically includes the following steps: inputting a first current pulse along a first direction into the spin-orbit moment channel layer at a first time, and inputting a second current pulse along a second direction into the spin-orbit moment channel layer at a second time. By setting the current density, pulse width, and timing of the first and second current pulses, deterministic flipping of the magnetic moment within the free layer can be achieved without an external magnetic field. That is, this application can achieve field-free directional flipping of the vertical magnetic moment by injecting two current pulses with different timings into the spin-orbit moment channel layer. Both the first and second current pulses are in-plane currents.

[0051] To achieve field-free magnetic moment reversal, the input direction of the current pulse, the timing of the current pulse (determined based on the start and end times of the current pulse and the pulse width (the duration from the rising edge to the falling edge of the pulse)), and the intensity of the current pulse (i.e. the current density of the current pulse) all have special requirements. The function of the current pulse can be adaptively adjusted according to different settings, and the ultimate goal is to achieve deterministic reversal of the free layer magnetic moment.

[0052] For the input direction of the current pulses, the input direction of the first current pulse (i.e., the first direction) and the input direction of the second current pulse (i.e., the second direction) intersect (e.g., the first current pulse and the second current pulse are orthogonal currents), and the plane formed by the first direction and the second direction is perpendicular to the stacking direction of the magnetic tunnel junction and the spin-orbit moment channel layer. For example, as Figure 9 and Figure 10 As shown, the first direction and the second direction can be directions parallel to the x-axis and y-axis, respectively.

[0053] Regarding the timing of the current pulses, this application designs to inject a first current pulse and a second current pulse within a dual current injection time step, and the timing relationship between the first current pulse and the second current pulse can be inconsistent between any two time steps. This application does not specifically limit the length of the time step; it can be determined based on the timing of the first current pulse and the second current pulse. The timing of the first current pulse and the second current pulse within a dual current injection time step will be specifically explained below.

[0054] Within a dual-current injection time step, the start time of the second current pulse (i.e., the second time) is no earlier than the start time of the first current pulse (i.e., the first time), the end times of the first current pulse and the second current pulse are different, and their timing overlaps. This application refers to the timing overlap meaning that the time interval for inputting the first current pulse (determined by the start and end times of the first current pulse) and the time interval for inputting the second current pulse (determined by the start and end times of the second current pulse) partially coincide on the time axis (partial overlap includes one or more time points coinciding). In this application, the start time refers to the time when the input current pulse begins, and the end time refers to the time when the input current pulse stops. Therefore, in this application, the start time of the second current pulse represents the time when the second current pulse begins to be input into the spin-orbit-moment channel layer, and the end time of the second current pulse represents the time when the second current pulse stops being input into the spin-orbit-moment channel layer. That is, this application requires that the first current pulse and the second current pulse within the same time step follow the following timing strategy: when the first time is equal to the second time, the end time of the first current pulse can be earlier or later than the end time of the second current pulse; when the first time is earlier than the second time, the end time of the first current pulse can be earlier than the end time of the second current pulse, and the first current pulse and the second current pulse have at least one overlapping time point on the time axis (i.e., there may be a time point at which the first current pulse and the second current pulse are input simultaneously); when the first time is earlier than the second time, the end time of the first current pulse can be later than the end time of the second current pulse, and in this case, the first current pulse and the second current pulse have multiple overlapping time points on the time axis.

[0055] More specifically, the timing relationship between the first current pulse and the second current pulse may exist. Figure 6 The diagram shows five timing scenarios, where the solid line segment represents the timing of the first current pulse and the dashed line segment represents the timing of the second current pulse. Example 1 indicates that the first time is equal to the second time, and the end time of the first current pulse is earlier than the end time of the second current pulse. Example 2 indicates that the first time is equal to the second time, and the end time of the first current pulse is later than the end time of the second current pulse. Example 3 indicates that the first time is earlier than the second time, and the end time of the first current pulse is later than the end time of the second current pulse. Examples 4 and 5 both indicate that the first time is earlier than the second time, and the end time of the first current pulse is earlier than the end time of the second current pulse. In Example 4, the first and second current pulses may have multiple overlapping time points on the time axis. In Example 5, the first and second current pulses have only one overlapping time point on the time axis. Figure 6 The examples in the text only represent the timing between the first and second current pulses, and do not represent the specific time points of the input current pulses.

[0056] As an example, this application mainly addresses the case where a second current pulse is input before the first current pulse has ended (i.e., there is an overlapping input period between the first and second current pulses). Therefore, the dual current injection method proposed in this application imposes the following requirement on the start time of the second current pulse: the start time of the second current pulse... [The start time of the first current pulse, the end time of the first current pulse], that is, the second time is no earlier than the start time of the first current pulse and no later than the end time of the first current pulse.

[0057] Furthermore, in the dual current injection method proposed in this application, the timing and current density settings of the current pulses affect the effect of the current pulses on the free-layer magnetic moment. Moreover, to more clearly illustrate the application of the dual current injection method proposed in this application to SOT devices in magnetic storage and Boolean logic operations, the following description will combine the magnetic storage function and Boolean logic function of SOT devices. It should be noted that a write operation can be achieved after one deterministic flip of the magnetic moment, and a Boolean logic operation can be achieved after one or more deterministic flips.

[0058] As an example, for the write function of a spin orbital moment device, only one time step is required to complete the write operation; for the Boolean logic function of a spin orbital moment device, the number of time steps is determined according to the entropy behavior of the logic operation. For example, XOR and XNOR logic operations correspond to three time steps, AND and OR logic operations correspond to two time steps, and TRUE and FALSE logic operations correspond to one time step. Therefore, the first time mentioned in this application can refer to the time interval of a single first current pulse, or it can refer to the time interval of the first current pulse within multiple time steps. Similarly, the second time can refer to the time interval of a single second current pulse, or it can refer to the time interval of the first current pulse within multiple time steps.

[0059] Regarding the current density of the current pulses, the vector sum of the current densities of the first and second current pulses is greater than the critical reversal current density, and the current density of either the first or second current pulse is less than the critical reversal current density. Specifically, under a first set condition, the vector sum of the current densities of the first and second current pulses is greater than the critical reversal current density, and the current density of the second current pulse is less than the critical reversal current density; under a second set condition, the vector sum of the current densities of the first and second current pulses is greater than the critical reversal current density, and the current density of the first current pulse is less than the critical reversal current density; under a third set condition, the current density of the first current pulse is greater than the critical reversal current density, and the current density of the second current pulse is less than the critical reversal current density. Regarding the function of the current pulses, the first and second current pulses are used to cause a perpendicular-in-plane reversal of the magnetization orientation of the magnetic moments within the free layer, and to deterministically convert the magnetization orientation of the transformed magnetic moments to a positive or negative polarity direction, thereby achieving a deterministic reversal of the magnetic moments within the free layer without an external magnetic field.

[0060] exist Figure 6 In Example 1, the first and second current pulses are simultaneously input into the spin-orbit moment channel layer, with the duration of the first current pulse being shorter than that of the second. Under this current injection, the current density vectors of the simultaneously input first and second current pulses can be superimposed to cause the magnetic moment to shift to an in-plane orientation. After the first current pulse is removed, the second current pulse, input alone, can change the magnetic moment orientation to a positive or negative polarity. The synergistic effect in Example 1 also applies to the current pulse input method in Example 2; that is, the sum of the current density vectors formed by the simultaneously input first and second current pulses can cause a perpendicular-to-in-plane shift in the magnetic moment, and the first current pulse alone can change the magnetic moment orientation to a positive or negative polarity. Figure 6 In Example 3, the vector sum of the current densities formed by the first and second current pulses can cause a vertical-in-plane transition in the magnetic moment. The first current pulse alone can change the magnetic moment orientation to positive or negative. In Example 4, the first current pulse (which may also include a second current pulse with an overlapping period) can cause the magnetic moment to change to an in-plane orientation. After the first current pulse is removed, the second current pulse alone can definitively reverse the magnetic moment to a positive or negative orientation. In Example 5, the first current pulse can cause a vertical-in-plane transition in the magnetic moment within the free layer, and the second current pulse can convert the changed magnetic moment to a positive or negative orientation.

[0061] Therefore, for the magnetic storage function (i.e., write function) of SOT devices, the functions performed by the first current pulse and the second current pulse are as follows: If the first current pulse and the second current pulse overlap only at one time point on the time axis, the first current pulse is used to cause a vertical-in-plane change in the magnetization orientation of the magnetic moment in the free layer, and the second current pulse is used to deterministically change the magnetization orientation of the magnetic moment in the free layer to a positive or negative polarity direction; if the first current pulse and the second current pulse overlap at multiple time points on the time axis, the first current pulse and the second current pulse are used together to cause a vertical-in-plane change in the magnetization orientation of the magnetic moment in the free layer, and if the end time of the second current pulse is later than the end time of the first current pulse, the second current pulse is also used to deterministically change the magnetization orientation of the magnetic moment in the free layer to a positive or negative polarity direction; if the end time of the second current pulse is earlier than the end time of the first current pulse, the first current pulse is also used to deterministically change the magnetization orientation of the magnetic moment in the free layer to a positive or negative polarity direction.

[0062] For the Boolean logic function of SOT devices, when the current density of the first current pulse and the second current pulse meet the current density settings under the first, second, or third setting conditions, if the magnetic moment component in the stacking direction of the magnetic tunnel junction is used as the logic output (different orientations of the magnetic moment component in the vertical direction correspond to different logic states; assuming that the vertical magnetic moment pointing upwards corresponds to the first Boolean logic state, and the vertical magnetic moment pointing downwards corresponds to the second Boolean logic state), the functions executed by the first current pulse and the second current pulse are as follows: If the logic output obtained based on the first and second current pulses is a first Boolean logic state or a second Boolean logic state, and the first and second current pulses overlap by only one time point on the time axis, then the first current pulse is used to cause a vertical-in-plane reversal of the magnetization orientation of the magnetic moments in the free layer, and the second current pulse is used to deterministically convert the magnetization orientation of the magnetic moments in the free layer to a positive or negative polarity direction (the output state of the SOT device can be determined according to the logic states corresponding to the first and second current pulses); if the logic output obtained based on the first and second current pulses is a first Boolean logic state or a second Boolean logic state... In a Boolean logic state, if the first current pulse and the second current pulse overlap at multiple time points on the time axis, then the first current pulse and the second current pulse are used together to cause a vertical-in-plane change in the magnetization orientation of the magnetic moment in the free layer. If the end time of the second current pulse is later than the end time of the first current pulse, the second current pulse is also used to deterministically change the magnetization orientation of the magnetic moment in the free layer to a positive or negative polarity direction. If the end time of the second current pulse is earlier than the end time of the first current pulse, the first current pulse is also used to deterministically change the magnetization orientation of the magnetic moment in the free layer to a positive or negative polarity direction.

[0063] In some embodiments of the present invention, for the writing function of the spin orbital moment device, the first setting condition includes a first time within the same time step being equal to a second time, or the first setting condition includes a first time within the same time step being equal to a second time and the end time of the second current pulse being later than the end time of the first current pulse (e.g., ...). Figure 6 Example 1); for the writing function of a spin orbital moment device, the second setting condition includes, within the same time step, the first time is equal to the second time and the end time of the second current pulse is earlier than the end time of the first current pulse, or, the first time is earlier than the second time and the end time of the second current pulse is earlier than the end time of the first current pulse (e.g. Figure 6 (Examples 2 and 3 in the text); for the writing function of the spin orbital moment device, the third setting condition includes a first time within the same time step being earlier than the second time, and the end time of the second current pulse within that time step being later than the end time of the first current pulse (as in Examples 4 and 5). In the first and second setting conditions, this application does not specifically set another unspecified current density magnitude; it is sufficient that the vector sum of the current densities of the first and second current pulses is greater than the critical flip current density.

[0064] The critical reversal current density is the threshold current density that induces a vertical magnetization reversal based on the spin-orbit moment. The critical reversal current density can be determined using the following formula: ; in, This refers to an externally applied in-plane magnetic field, and this application does not need to consider... In this case, the critical reversal current density is the critical current density that can pull the magnetic moment into the plane.

[0065] For the Boolean logic function of the spin orbital moment device, the first setting condition may include that within the same time step, the first time is equal to the second time, and the end time of the second current pulse is later than the end time of the first current pulse (corresponding to...). Figure 6 (The timing relationship shown in Example 1) and the logic state corresponding to the first current pulse is the first Boolean logic state; for the Boolean logic function of the spin orbital moment device, the second setting condition may include that the first time is equal to the second time within the same time step, and the end time of the second current pulse is earlier than the end time of the first current pulse (corresponding to...). Figure 6 (The timing relationship in Example 2) and the logic state corresponding to the first current pulse is the first Boolean logic state, or, within the same time step, the first time is earlier than the second time, and the end time of the second current pulse is earlier than the end time of the first current pulse (corresponding to...) Figure 6(The timing relationship shown in Example 3) and the logic state corresponding to the first current pulse is the first Boolean logic state; the third setting condition may include, within the same time step, the first time being earlier than the second time and the end time of the second current pulse being later than the end time of the first current pulse (corresponding to...) Figure 6 (The timing relationship is shown in Examples 4 and 5), and the logic state corresponding to the first current pulse is the first Boolean logic state.

[0066] It should be noted that, Figure 6 In the timing relationship shown in Example 4, based on the length of overlap between the time intervals of the first and second current pulses, it can be divided into the following two cases: ① The current densities of both the first and second current pulses are less than the critical reversal current density, and the vector sum of the current densities of the first and second current pulses is greater than the critical reversal current density; ② The current density of the first current pulse is greater than the critical reversal current density, and the current density of the second current pulse is less than the critical reversal current density (in this case, the vector sum of the current densities of the first and second current pulses must be greater than the critical reversal current density). To achieve reversal, in case ① of Example 4, the number of overlapping time points of the first and second current pulses on the time axis must be greater than the number of overlapping time points in case ②. That is, the current density setting in case ② does not need to consider the influence of the overlap time, while case ① needs to consider the overlap time. This application does not specifically limit the overlap time of the time intervals of the first and second current pulses in case ① of Example 4, and the overlap time can be determined by simulation and other means, which will not be described in detail here. To achieve deterministic magnetic moment reversal, this application only specifically describes the current injection situation in case ② of Example 4 (in which case the vector sum of the current density of the first current pulse and the second current pulse is greater than the critical reversal current density).

[0067] This application uses "first Boolean logic state" and "second Boolean logic state" to represent two relative logic states in Boolean logic operations. For example, Boolean values ​​1 and 0 can be used to represent these two logic states respectively. Hereinafter, 1 is used to represent the first Boolean logic state and 0 is used to represent the second Boolean logic state.

[0068] As an example, for the Boolean logic function of a spin orbital moment device, when the logic state corresponding to the first current pulse is the second Boolean logic state, the current densities of the first and second current pulses can be set based on a set current strategy. The set current strategy includes: If the first and second current pulses overlap multiple times on the time axis within the same time step, the vector sum of their current densities within that time step is less than the critical switching current density (in this case, the current densities of both the first and second current pulses are less than the critical switching current density). If the first and second current pulses overlap only once on the time axis within the same time step, then the current densities of both the first and second current pulses are less than the critical switching current density. That is, the current strategy is set such that the current densities of both the first and second current pulses are less than the critical switching current density.

[0069] As an example, the input direction when the first current pulse corresponds to the first Boolean logic state is on the same straight line as the input direction when the first current pulse corresponds to the second Boolean logic state, and the input direction when the second current pulse corresponds to the first Boolean logic state is opposite to the input direction when the second current pulse corresponds to the second Boolean logic state. For example, as... Figure 11 As shown, when the logic state of the first current pulse is the first Boolean logic state and the second Boolean logic state, the input direction is +x (or reversed: when the logic state of the first current pulse is the first Boolean logic state, the input direction can be +x; when the logic state of the first current pulse is the second Boolean logic state, the input direction can be -x). When the logic state of the second current pulse is the first Boolean logic state, the input direction of the second current pulse can be +y (when the logic state is the second Boolean logic state, the input direction of the second current pulse can be -y).

[0070] In some embodiments of the present invention, the minimum pulse width required for the first current pulse and the second current pulse can be achieved through simulation. For example, given the current density of the first current pulse and the current density of the second current pulse, the pulse width that can deterministically reverse the free layer magnetic moment can be determined using simulation software such as MATLAB (i.e., the pulse widths of the first current pulse and the second current pulse can be determined based on the current density of the first current pulse and the current density of the second current pulse through simulation). Figure 6 Taking the timing relationship in Example 5 as an example, assume the current density of the first current pulse is... The current density of the second current pulse is If the pulse width of the first current pulse The value of and the pulse width of the second current pulse The intersection of the values ​​is located at Figure 7 Within the first region (blue region), based on the timing relationship in Example 5, the current density of the first and second current pulses, and the determined pulse width, the first and second current pulses can be input to achieve a deterministic reversal of the free layer magnetic moment. The values ​​and The intersection of the values ​​is located at Figure 7 Within the second region (green region), deterministic flipping cannot be achieved if... The values ​​and The intersection of the values ​​is located at Figure 7 Within the third region (yellow region), the free layer magnetic moment cannot be flipped.

[0071] Figure 8 This is a magnetization dynamics trajectory diagram obtained from macro-spin simulation in a Ta / CoFeB / MgO / CoFeB system (in order: SOT channel layer / free layer / barrier layer / reference layer). The current density of the first current pulse is 0.2 times the critical flip current density, and the current density of the second current pulse is 0.6 times the critical flip current density. With both the first and second current pulses having a pulse width of 1 millisecond, the following can be obtained: Figure 8 The magnetization dynamics trajectory diagram is shown.

[0072] by Figure 6 Taking the timing case of Example 5 as an example, the principle of the dual current injection method proposed in this application can be analyzed as follows: Suppose that two current pulses are passed through a magnetic storage device that has a cross-shaped top view. The first current pulse is a current pulse J in the x-axis direction. x The second current pulse is a current pulse J in the y-axis direction. y Both pulse currents can produce damped SOT and field-like SOT, and as... Figure 9 As shown in (b), the end time of the first current pulse is the same as the start time of the second current pulse (using...). Figure 6 The timing relationship is shown in Example 5. The principle of magnetic moment reversal during dual current injection can be described as follows: Figure 10 As shown, Figure 10 The gray arrow indicates the direction of the free-layer magnetic moment (the magnetic moment is perpendicular when the current is first applied), and the green arrows indicate the direction of the free-layer magnetic moment. and damping SOT The yellow arrows represent the equivalent fields of the field class. and damping equivalent field .like Figure 10 As shown in (a), the gray arrow represents the initial current pulse J. x When inputting in the +x direction, the direction of the magnetic moment is such that when the current density of the current pulse in the +x direction is greater than the critical reversal current density, the current generated along the x-axis... and Under the influence of this, the magnetic moment will be pulled towards the +y direction (as shown by the blue arrow in the polarization direction of the spin current injected into the free layer). Figure 10 As shown in (b), the gray arrow represents the initial current pulse J. yThe direction of the magnetic moment when inputting in the +y direction.

[0073] Initially, J was introduced. y At that time, J y The free layer produced The symmetry is broken in the -z direction, at which point the magnetic moment flips to the lower hemisphere. Continue introducing J... y The magnetic moment can achieve field-free deterministic reversal of the perpendicular magnetic moment. That is, the above utilizes orthogonal current J x and J y When achieving field-free switching, the first current pulse J x Its function is to pull the magnetic moment into the plane, and the second current pulse J y Its function is to generate a downward torque, thereby breaking the symmetry and causing the magnetic moment to flip down to the lower hemisphere.

[0074] In some specific embodiments of the present invention, based on the dual-current field-free switching mechanism, the x-axis current J can be... x and y-axis current J y Two input variables are used to perform logical operations. Assuming the Boolean value corresponding to the first Boolean logic state is 1 and the Boolean value corresponding to the second Boolean logic state is 0, this application injects a first current pulse J into the SOT device. x Second current pulse J y It can achieve Figure 11 The 16 Boolean logic operations are shown in (b) of the paper. Furthermore, to implement multiple Boolean logic operations, this application defines the Boolean logic states corresponding to the first and second current pulses as follows: When the logic state corresponding to the first current pulse (hereinafter referred to as character A) is the second Boolean logic state (A=0), regardless of whether the logic state corresponding to the second current pulse (hereinafter referred to as character B) is the first or second Boolean logic state (B=0 or B=1), the logic output of the SOT device (taking the magnetic moment component in the magnetic tunnel junction stacking direction as the logic output) is the same as the logic state corresponding to the magnetic moment before the input current pulse at that time step (the magnetic moment direction remains unchanged); when the logic state corresponding to the first current pulse is the first Boolean logic state (A=1), if the logic state corresponding to the second current pulse is the first Boolean logic state (B=1), the logic output of the SOT device is always the first Boolean logic state (regardless of the initial magnetization state); if the logic state corresponding to the second current pulse is the second Boolean logic state (B=0), the logic output of the SOT device is always the second Boolean logic state (regardless of the initial magnetization state).

[0075] Furthermore, the current density corresponding to the first current pulse (and the second current pulse) can be determined based on the logic state and timing relationship of the first current pulse (and the second current pulse). For Figure 6The timing relationship shown in Example 1, when A=0 (regardless of whether B=0 or B=1), J x and J y The vector sum is less than the critical flip current density When A=1 (regardless of whether B=0 or B=1), J x and J y The vector sum is greater than the critical reversal current density. And J y The value of is always less than For example, the correspondence between the logic state and current density of the first and second current pulses can be: when A=0, J x When J = 0, A = 1 x = When B=0, J y =0.6 And J when B=1 y =-0.6 .for Figure 6 The timing relationship shown in Example 2 indicates that when A=0 (regardless of whether B=0 or B=1), J x and J y The vector sum is less than When A=1 (regardless of whether B=0 or B=1), J x and J y The vector sum is greater than the critical reversal current density. And J x The value of is always less than For example, the correspondence between the logic state and current density of the first and second current pulses can be: when A=0, J x When J = 0, A = 1 x =0.6 When B=0, J y =-0.6 And J when B=1 y =0.6 .for Figure 6 The timing relationship shown in Example 3 indicates that when A=0 (regardless of whether B=0 or B=1), J x and J y The vector sum is less than When A=1 (regardless of whether B=0 or B=1), J x and J y The vector sum is greater than And J x The value of is always less than For example, the correspondence between the logic state and current density of the first and second current pulses can be: when A=0, J x When J = 0, A = 1 x=0.6 When B=0, J y =-0.6 And J when B=1 y =0.6 .for Figure 6 The timing relationship shown in Example 4 indicates that when A=0 (regardless of whether B=0 or B=1), J x and J y The vector sum is less than When A=1 (regardless of whether B=0 or B=1), J x and J y The vector sum is greater than And J y The value is less than For example, the correspondence between the logic state and current density of the first and second current pulses can be: when A=0, J x When J = 0, A = 1 x = When B=0, J y =0.6 And J when B=1 y =-0.6 .for Figure 6 The time series relationship shown in Example 5, when A=0, J x The value is less than When A=1, J x The value is greater than And J y The value of is always less than .

[0076] As an example, this application can convert the z-direction component m of the magnetic moment of the ferromagnetic layer. z As a logical output, it identifies the magnetic moment upward state (logic "1") and the magnetic moment downward state (logic "0"). Based on the above correspondence between current density and logical state, a complete set of 16 Boolean logic operations can be implemented. For specific logic operation rules, see [link to relevant documentation]. Figure 11 As shown in (b), p and q are two input logic variables, and "-" indicates no operation. " represents negation, that is 0 = 1, 1 = 0.

[0077] by Figure 6 Taking the time sequence relationship shown in Example 5 as an example, Figure 11 Tables (c)-(e) show the simulation results of Boolean logic functions based on the induction of flipping by orthogonal currents (the first and second directions are perpendicular). Figure 11As shown in (c), when performing the p logical operation, if the input p = 1 and q = 0, then the logical operation "A = 1, B = p = 1" needs to be performed (for example, passing J along the x-axis). x = Then J is inserted along the y-axis y = -0.6 The final magnetized state is upward, and the output logic value is 1. Regardless of the input values ​​for p and q, the logic operation on p can be completed in one step. For example... Figure 11 In (d), for the AND logic operation, when the logic input p=q=0, an initialization step (step 1) is required first, executing the logic operation "A=1, B=0" (at this time, J can be passed along the x-axis). x = J is inserted along the y-axis y = 0.6 The second step (step 2) involves performing the logical operation "A=p=0, B=q=0" (for example, inserting J along the x-axis). x = 0.6 And pass J along the y-axis y = 0.6 This ultimately causes the magnetic moment to flip downwards, outputting a logic value of 0. For other "pq" combinations ("01", "10", and "11"), first execute the initialization step (step 1) "A= 1, B= 0", then execute step 2 "A=p, B=q" to achieve the logic function. Furthermore, the XNOR logic operation can be implemented in three steps. When the logic input p= 0, q= 0, the first step is to execute the "A= 1, B= 0" logic operation (at this time, J can be passed along the x-axis). x = And pass J along the y-axis y = 0.6 The second step is to execute the logical operation "A=p=0, B=q=0" (passing J along the x-axis). x = 0.6 J is inserted along the y-axis y = 0.6 The third step is to execute "A= p=1, B= q = 1” logical operation (passing J along the x-axis) x = J is inserted along the y-axis y = -0.6 After these three steps, the final magnetized state faces upward, and the output logic value is 1, as shown below. Figure 11As shown in (e). For other "pq" combinations ("01", "10", and "11"), the first step is to execute "A= 1, B= 0", the second step is to execute "A=p, B=q", and the third step is to execute "A= p, B= The logical function can be achieved by using "q".

[0078] Furthermore, this invention designs and simulates to verify that by using orthogonal currents as input variables and employing a step-by-step input method, all 16 Boolean logic functions can be implemented in a single device. This switching mechanism, which requires no external magnetic field and provides a complete Boolean logic implementation scheme, is fully compatible with the widely used Ta (or W) / CoFeB / MgO / CoFeB magnetic tunnel junction structure in the industry, and holds promise for combining non-volatile memory and computing units in integrated circuits in the future, realizing in-memory computing devices.

[0079] The dual current injection method proposed in this application has the following significant advantages: (1) The dual-current input method proposed in this application can use existing SOT devices to realize the magnetic moment orientation reversal induced by the spin-orbit moment effect, thereby realizing the magnetic moment reversal of the free layer without the assistance of an external magnetic field; (2) The SOT devices applicable to this application can use a wide range of magnetic tunnel junction materials, thereby realizing field-free reversal by utilizing widely used magnetic tunnel junction (e.g., Ta / CoFeB / MgO / CoFeB) structures, which is easy to match with existing processes and facilitates large-scale promotion; (3) SOT devices have vertical magnetic anisotropy, which can improve storage density and thermal stability; (4) The tunnel junction structure used in this invention is simple and does not require the introduction of additional crystal structures. The integration process is simple and has few steps, which can reduce process complexity and manufacturing cost, and is compatible with current large-scale integrated circuit processes; (5) All 16 Boolean logics can be implemented in a single SOT device, providing a new path for the realization of in-memory computing.

[0080] Corresponding to the above method, the present invention also provides a dual-current magnetic moment reversing device for a spin orbital moment device, the device comprising a driver and a means for generating current pulses, the driver being used to drive the means for generating current pulses to perform the steps of the method as described above.

[0081] Those skilled in the art will understand that the exemplary components, systems, and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Whether implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention. When implemented in hardware, it can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this invention are programs or code segments used to perform the desired tasks. The programs or code segments can be stored in a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried in a carrier wave.

[0082] It should be clarified that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the present invention.

[0083] In this invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or in place of features of other embodiments.

[0084] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, various modifications and variations of the embodiments of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for dual-current magnetic moment reversal in a spin orbital moment device, characterized in that, The spin orbital moment device has perpendicular magnetic anisotropy and includes a magnetic tunnel junction containing a free layer and a spin orbital moment channel layer stacked below the free layer; the method includes the following steps: A first current pulse is input into the spin-orbit-moment channel layer along a first direction at a first time, and a second current pulse is input into the spin-orbit-moment channel layer along a second direction at a second time, wherein the first direction and the second direction intersect, and the plane defined by the first direction and the second direction is perpendicular to the stacking direction of the magnetic tunnel junction and the spin-orbit-moment channel layer. Within a dual-current injection time step, the first time is earlier than or equal to the second time, the timing of the first current pulse and the second current pulse overlaps, and the end times of the first current pulse and the second current pulse are different; the vector sum of the current densities of the first current pulse and the second current pulse is greater than the critical reversal current density, and the current density of either the first current pulse or the second current pulse is less than the critical reversal current density; furthermore, the first current pulse and the second current pulse are used to cause a vertical-in-plane reversal of the magnetization orientation of the magnetic moments in the free layer, and to deterministically convert the magnetization orientation of the transformed magnetic moments to a positive or negative polarity direction, so that a deterministic reversal of the magnetic moments in the free layer can be achieved without an external magnetic field.

2. The method according to claim 1, characterized in that, The vector sum of the current densities of the first current pulse and the second current pulse is greater than the critical reversal current density, and the current density of either the first current pulse or the second current pulse is less than the critical reversal current density, including: Under a first set condition, the vector sum of the current densities of the first current pulse and the second current pulse is greater than the critical reversal current density, and the current density of the second current pulse is less than the critical reversal current density. Under the second set condition, the vector sum of the current densities of the first current pulse and the second current pulse is greater than the critical reversal current density, and the current density of the first current pulse is less than the critical reversal current density. Under the third set condition, the current density of the first current pulse is greater than the critical reversal current density, and the current density of the second current pulse is less than the critical reversal current density.

3. The method according to claim 2, characterized in that, For the writing function of the spin orbital moment device, the first setting condition includes that within one time step, the first time is equal to the second time and the end time of the second current pulse is later than the end time of the first current pulse. For the writing function of the spin orbital moment device, the second setting condition includes, within one time step, a first time equal to a second time and the end time of the second current pulse earlier than the end time of the first current pulse, or, a first time earlier than a second time and the end time of the second current pulse earlier than the end time of the first current pulse. For the writing function of the spin orbital moment device, the third setting condition includes that within one time step, the first time is earlier than the second time and the end time of the second current pulse is later than the end time of the first current pulse. For the Boolean logic function of the spin orbit moment device, the first setting condition includes that within one time step, the first time is equal to the second time, the end time of the second current pulse is later than the end time of the first current pulse, and the logic state corresponding to the first current pulse is the first Boolean logic state. For the Boolean logic function of the spin orbital moment device, the second setting condition includes, within one time step, the first time is equal to the second time, the end time of the second current pulse is earlier than the end time of the first current pulse and the logic state corresponding to the first current pulse is the first Boolean logic state, or, the first time is earlier than the second time, the end time of the second current pulse is earlier than the end time of the first current pulse and the logic state corresponding to the first current pulse is the first Boolean logic state. For the Boolean logic function of the spin orbit moment device, the third setting condition includes that within one time step, the first time is earlier than the second time, the end time of the second current pulse is later than the end time of the first current pulse, and the logic state corresponding to the first current pulse is the first Boolean logic state.

4. The method according to claim 2, characterized in that, Regarding the Boolean logic function of the spin orbital moment device, when the logic state corresponding to the first current pulse is the second Boolean logic state, the current densities of the first and second current pulses are set based on a predetermined current strategy; wherein, the predetermined current strategy includes: If the first current pulse and the second current pulse overlap multiple time points on the time axis within a time step, the vector sum of the current density of the first current pulse and the second current pulse is less than the critical flip current density. If the first current pulse and the second current pulse overlap by only one time point on the time axis within a time step, then the current density of the first current pulse and the current density of the second current pulse are both less than the critical flip current density.

5. The method according to claim 4, characterized in that, The input direction when the first current pulse corresponds to the first Boolean logic state is on a straight line with the input direction when the first current pulse corresponds to the second Boolean logic state, and the input direction when the second current pulse corresponds to the first Boolean logic state is opposite to the input direction when the second current pulse corresponds to the second Boolean logic state.

6. The method according to claim 1, characterized in that, The first current pulse and the second current pulse perform the following functions respectively: If the first current pulse and the second current pulse overlap by only one time point on the time axis, the first current pulse is used to cause a vertical-in-plane reversal of the magnetization orientation of the magnetic moment in the free layer, and the second current pulse is used to definitively change the magnetization orientation of the magnetic moment in the free layer to a positive or negative polarity. Otherwise, the first current pulse and the second current pulse are used together to cause a vertical-in-plane reversal of the magnetization orientation of the magnetic moment in the free layer. If the end time of the second current pulse is later than the end time of the first current pulse, the second current pulse is used to definitively change the magnetization orientation of the magnetic moment in the free layer to a positive or negative polarity. If the end time of the second current pulse is earlier than the end time of the first current pulse, the first current pulse is used to definitively change the magnetization orientation of the magnetic moment in the free layer to a positive or negative polarity.

7. The method according to claim 1, characterized in that, The spin orbit moment device further includes a first current axis layer and a second current axis layer respectively connected to the spin orbit moment channel layer, the first current axis layer and the second current axis layer being located above the substrate layer; wherein, the first current axis layer is used to inject a first current pulse into the spin orbit moment channel layer, and the second current axis layer is used to inject a second current pulse into the spin orbit moment channel layer.

8. The method according to claim 1, characterized in that, The top view of the spin orbital moment device is cross-shaped or star-shaped.

9. A spin orbital moment device, characterized in that, The spin-orbit moment device has vertical magnetic anisotropy and includes: a top electrode, a magnetic tunnel junction located below the top electrode and containing a free layer, a spin-orbit moment channel layer located below the free layer, a base layer located below the spin-orbit moment channel layer, and a first current axis layer and a second current axis layer respectively connected to the spin-orbit moment channel layer; the first current axis layer and the second current axis layer are located above the base layer, the first current axis layer is used to inject a first current pulse into the spin-orbit moment channel layer, and the second current axis layer is used to inject a second current pulse into the spin-orbit moment channel layer; The spin orbital moment device is used to achieve a deterministic reversal of the magnetic moment within the free layer by the method described in any one of claims 1-8.

10. A dual-current magnetic moment reversal device for a spin orbital moment device, characterized in that, The device includes a driver and a means for generating current pulses, the driver being used to drive the means for generating current pulses to perform the steps of the method as described in any one of claims 1 to 8.

Images