Domain wall displacement element and magnetic array

Abstract

A domain wall displacement element according to the present invention comprises a reference layer, a non-magnetic layer, and a domain wall displacement layer. The non-magnetic layer is laminated on the reference layer and is sandwiched, in the lamination direction, between the reference layer and the domain wall displacement layer. The reference layer is provided with a first ferromagnetic layer, an antiferromagnetic coupling layer, and a second ferromagnetic layer. In the lamination direction, the antiferromagnetic coupling layer is sandwiched between the first ferromagnetic layer and the second ferromagnetic layer. The side surface of the reference layer is inclined with respect to the lamination direction. The relationship 3 × W10 < W30 < L10 is satisfied, where L10 is the first-direction length of the domain wall displacement layer, the first direction being the longitudinal direction of the domain wall displacement layer, W10 is the second-direction width of a first surface of the domain wall displacement layer that is on the side further from the non-magnetic layer, the second direction being a direction orthogonal to both the lamination direction and the first direction, and W30 is the second-direction width of a second surface of the reference layer that is on the side further from the non-magnetic layer.

Description

Domain wall motion element and magnetic array

[0001] The present disclosure relates to domain wall motion elements and magnetic arrays.

[0002] Magnetoresistive effect elements are known that utilize a change in resistance value (magnetoresistance change) based on a change in the relative angle between the magnetizations of two ferromagnetic layers. For example, a domain wall motion type magnetoresistive effect element (hereinafter referred to as a domain wall motion element) described in Patent Document 1 is an example of a magnetoresistive effect element. In a domain wall motion element, the resistance value in the stacking direction changes depending on the position of the domain wall, and data can be recorded in a multi-value or analog format. A domain wall motion element has high linearity and symmetry in the resistance change, excellent rewrite durability, and is capable of high-speed operation.

[0003] An example of a domain wall motion element is disclosed in Patent Document 1. Patent Document 1 discloses that the relationship between the film thickness and saturation magnetization of the magnetic recording layer is specified in order to reduce the influence of a leakage magnetic field on other elements.

[0004] International Publication No. 2021 / 166892

[0005] The leakage magnetic field from the reference layer of the domain wall motion element may affect the domain wall motion layer in addition to other elements, and the leakage magnetic field applied to the domain wall motion layer may disturb the motion of the domain wall.

[0006] The present disclosure has been made in consideration of the above problems, and aims to provide a domain wall motion element and a magnetic array that can be easily manufactured and that can reduce the leakage magnetic field applied to the domain wall motion layer.

[0007] A domain wall motion element according to a first aspect includes a reference layer, a nonmagnetic layer, and a domain wall motion layer. The nonmagnetic layer is stacked on the reference layer and sandwiched between the reference layer and the domain wall motion layer in the stacking direction. The reference layer includes a first ferromagnetic layer, an antiferromagnetic coupling layer, and a second ferromagnetic layer. In the stacking direction, the antiferromagnetic coupling layer is sandwiched between the first ferromagnetic layer and the second ferromagnetic layer. A side surface of the reference layer is inclined with respect to the stacking direction. When the longitudinal direction of the domain wall motion layer is defined as a first direction, the direction perpendicular to the stacking direction and the first direction is defined as a second direction, the length of the domain wall motion layer in the first direction is defined as L10, the width in the second direction of a first surface of the domain wall motion layer farther from the nonmagnetic layer is defined as W10, and the width in the second direction of a second surface of the reference layer farther from the nonmagnetic layer is defined as W30, the relationship 3×W10<W30<L10 is satisfied.

[0008] The domain wall motion element and magnetic array according to the present disclosure can be easily manufactured, and can reduce the leakage magnetic field applied to the domain wall motion layer.

[0009] FIG. 1 is a block diagram of a magnetic array according to the first embodiment. FIG. 2 is a circuit diagram of an integrated region of the magnetic array according to the first embodiment. FIG. 3 is a cross-sectional view of the vicinity of a domain wall motion element of the magnetic array according to the first embodiment. FIG. 4 is a plan view of a domain wall motion element according to the first embodiment. FIG. 5 is a cross-sectional view of a domain wall motion element according to the first embodiment. FIG. 6 is another cross-sectional view of a domain wall motion element according to the first embodiment. FIG. 7 is a simulation result of a leakage magnetic field applied to a domain wall motion layer. FIG. 8 is a schematic view of a model used in the simulation of the leakage magnetic field. FIG. 9 is a conceptual view of a neural network. FIG. 10 is a block diagram showing a system including a neuromorphic device according to the first embodiment. FIG. 11 is a cross-sectional view of a domain wall motion element according to the second embodiment. FIG. 12 is a cross-sectional view of a domain wall motion element according to the third embodiment. FIG. 13 is a cross-sectional view of a domain wall motion element according to the fourth embodiment. FIG. 14 is a plan view of a domain wall motion element according to the fifth embodiment. FIG. 15 is a plan view of a domain wall motion element according to the sixth embodiment.

[0010] The present embodiment will be described in detail below with reference to the accompanying drawings. The drawings used in the following description may show characteristic portions enlarged for the sake of clarity, and the dimensional ratios of the components may differ from the actual values. The materials, dimensions, and the like exemplified in the following description are merely examples, and the present embodiment is not limited thereto. Appropriate modifications can be made within the scope of the effects of the present embodiment.

[0011] First, the directions will be defined. The x-direction and y-direction are directions substantially parallel to one surface of the substrate Sub (see FIG. 3 ), which will be described later. The x-direction is the longitudinal direction of the first ferromagnetic layer, which will be described later. The x-direction is an example of a first direction. The y-direction is a direction perpendicular to the x-direction when viewed from the stacking direction. The y-direction is an example of a second direction. The z-direction is a direction from the substrate, which will be described later, to the domain wall motion element. The z-direction is an example of a stacking direction. In this specification, the +z direction may be expressed as "up" and the -z direction as "down," but these expressions are for convenience and do not define the direction of gravity. Furthermore, in this specification, "extending in the x-direction" means, for example, that the dimension in the x-direction is larger than the smallest dimension among the dimensions in the x-direction, y-direction, and z-direction. The same applies to extending in other directions.

[0012] 1 is a block diagram of a magnetic array MA according to a first embodiment. The magnetic array MA has an integration region 1 and a peripheral region 2. The magnetic array MA can be used, for example, in a magnetic memory, a multiply-and-accumulate unit, a neuromorphic device, a spin memristor, or a magneto-optical element.

[0013] The accumulation region 1 is a region where a plurality of domain wall motion elements are accumulated. When the magnetic array MA is used as a memory, data is stored in the accumulation region 1. When the magnetic array MA is used as a neuromorphic device, learning and inference are performed in the accumulation region 1.

[0014] The peripheral region 2 is a region where a control element that controls the operation of the domain wall motion element in the integration region 1 is mounted. The peripheral region 2 includes, for example, a control device 3, a resistance detection device 4, and an output unit 5.

[0015] The control device 3 is configured to be able to apply a pulse to at least one of the plurality of domain wall motion elements in the accumulation region 1. The control device 3 includes, for example, a control unit 6 and a power supply 7.

[0016] The control unit 6 has, for example, a processor and a memory. The processor is, for example, a CPU (Central Processing Unit). The processor operates based on an operating program stored in the memory. The control unit 6 controls, for example, the address of the domain wall motion element to which the pulse is applied, the magnitude (voltage, pulse length) of the pulse to be applied to a specific domain wall motion element, etc. The control unit 6 may also have a clock, a counter, a random number generator, etc. The clock serves as an indicator of the timing to apply the pulse, and the counter counts the number of times the pulse is applied, etc. The power supply 7 applies pulses to the domain wall motion element according to instructions from the control unit 6.

[0017] The resistance detection device 4 is configured to detect the resistance value of the domain wall motion element in the integration region 1. The resistance detection device 4 may detect the resistance of each domain wall motion element in the integration region 1, or may detect the total resistance of domain wall motion elements belonging to the same column, for example. The resistance detection device 4 may have, for example, a comparator that compares the magnitude of the detected resistance value. The comparator may, for example, compare the detected resistance values ​​with each other, or may compare the detected resistance value with a preset reference resistance value.

[0018] The output unit 5 is connected to the resistance detection device 4. The output unit 5 includes, for example, a processor, an output capacitor, an amplifier, a converter, etc. When the magnetic array MA is used as a neuromorphic device, the output unit 5 may perform a calculation to substitute the detection result of the resistance detection device 4 into an activation function. The calculation is performed, for example, by a processor. The output unit 5 outputs the calculation result to the outside. When the magnetic array MA is used as a neuromorphic device, for example, the calculation result may be output as an input signal for another magnetic array, or may be output as a discrimination rate to the outside. The output unit 5 may also feed back the calculation result to the control device 3.

[0019] 2 is a circuit diagram of the integrated region 1 according to the first embodiment. The integrated region 1 includes a plurality of domain wall motion elements 100, a plurality of write wirings WL, a plurality of common wirings CL, a plurality of read wirings RL, a plurality of first switches SW1, and a plurality of second switches SW2. The third switch SW3 may belong to the control device 3 in the peripheral region 2, for example.

[0020] The plurality of domain wall motion elements 100 are arranged, for example, in a matrix. The plurality of domain wall motion elements 100 are not limited to elements that are actually arranged in a matrix, but may also be elements that are arranged in a matrix in a circuit diagram.

[0021] Each of the write wirings WL is used when writing data. Each of the write wirings WL electrically connects the control device 3 to one or more domain wall motion elements 100. Each of the common wirings CL is used when writing and reading data. Each of the common wirings CL is connected to, for example, the resistance detection device 4. The common wiring CL may be provided for each of the multiple domain wall motion elements 100, or may be provided across the multiple domain wall motion elements 100. Each of the read wirings RL is used when reading data. Each of the read wirings RL electrically connects the control device 3 to one or more domain wall motion elements 100.

[0022] The first switch SW1, the second switch SW2, and the third switch SW3 are elements that control the flow of current. The first switch SW1, the second switch SW2, and the third switch SW3 may be, for example, an element that uses a phase change of a crystal layer such as a transistor or an Ovonic Threshold Switch (OTS), an element that uses a change in band structure such as a Metal-Insulator Transition (MIT) switch, an element that uses a breakdown voltage such as a Zener diode or an avalanche diode, or an element that changes conductivity according to a change in atomic position.

[0023] The first switch SW1 and the second switch SW2 are connected, for example, one to each of the domain wall motion elements 100. The first switch SW1 is connected, for example, between the domain wall motion element 100 and the write wiring WL. The second switch SW2 is connected, for example, between the domain wall motion element 100 and the common wiring CL. The third switch SW3 is connected, for example, across a plurality of domain wall motion elements 100. The third switch SW3 is connected, for example, to the read wiring RL.

[0024] The positional relationship between the first switch SW1, the second switch SW2, and the third switch SW3 is not limited to that shown in Fig. 2. For example, the first switch SW1 may be connected across multiple domain wall motion elements 100 and located upstream of the write wiring WL. Also, for example, the second switch SW2 may be connected across multiple domain wall motion elements 100 and located upstream of the common wiring CL. Also, for example, the third switch SW3 may be connected to each domain wall motion element 100.

[0025] 3 is a cross-sectional view of the vicinity of the domain wall motion element 100 in the integration region 1 according to the first embodiment. Fig. 3 is a cross-section of one domain wall motion element 100 in Fig. 2 taken along an xz plane passing through the center of the width of the domain wall motion layer 10 in the y direction.

[0026] The first switch SW1 and the second switch SW2 shown in Figure 3 are transistors Tr. The transistor Tr has a gate electrode G, a gate insulating film GI, a source S, and a drain D. The source S and the drain D are determined by the direction of current flow and are both semiconductor active regions. Figure 3 shows only one example, and the positional relationship between the source S and the drain D may be reversed. The substrate Sub is, for example, a semiconductor substrate. The third switch SW3 is electrically connected to the readout wiring RL and is, for example, located at a position shifted in the y direction in Figure 3.

[0027] The transistor Tr, the write wiring WL, the common wiring CL, the read wiring RL, and the domain wall motion element 100 are connected by via wirings V extending in the z direction or in-plane wirings IP extending in any direction within the xy plane. The via wirings V and the in-plane wirings IP contain a conductive material. An insulating layer 90 is formed between different layers in the z direction, except for the via wirings V.

[0028] The insulating layer 90 is an insulating layer that insulates between wirings in a multilayer wiring structure and between elements. The domain wall motion element 100 and the transistor Tr are electrically isolated by the insulating layer 90 except for the via wiring V. The insulating layer 90 is made of, for example, silicon oxide (SiO ｘ ), silicon nitride (SiN ｘ ), silicon carbide (SiC), chromium nitride, silicon carbonitride (SiCN), silicon oxynitride (SiON), aluminum oxide (Al ２ O ３ ), zirconium oxide (ZrO ｘ ) etc.

[0029] FIG. 4 is a plan view of the domain wall motion element 100 as viewed from the z direction. FIG. 5 is a cross-sectional view of the domain wall motion element 100 cut along the xz plane passing through the center of the domain wall motion layer 10 in the y direction. FIG. 5 corresponds to the cut surface cut along line A-A in FIG. 4. FIG. 6 is a cross-sectional view of the domain wall motion element 100 cut along the yz plane passing through the center of the domain wall motion layer 10 in the x direction. FIG. 6 corresponds to the cut surface cut along line B-B in FIG. 4. The arrows shown in FIGS. 5 and 6 are examples of the orientation direction of the magnetization of a ferromagnetic material in the initial state when no external magnetic field is applied to the domain wall motion element 100.

[0030] The domain wall motion element 100 includes, for example, a domain wall motion layer 10, a non-magnetic layer 20, a reference layer 30, a first magnetization pinned layer 40, a second magnetization pinned layer 50, a first electrode E1, a second electrode E2, and a third electrode E3. Each of the multiple domain wall motion elements included in the accumulation region 1 is, for example, the domain wall motion element 100 according to this embodiment.

[0031] The domain wall displacement layer 10 extends in the x direction. When viewed from the z direction, the length of the domain wall displacement layer 10 in the x direction is longer than the length in the y direction. The x direction is the longitudinal direction of the domain wall displacement layer 10. The length L10 of the domain wall displacement layer 10 in the x direction is longer than the width W10 in the y direction. The length L10 and width W10 are the length and width of the first surface of the domain wall displacement layer 10 on the side farther from the nonmagnetic layer 20.

[0032] The domain wall displacement layer 10 has two magnetic domains therein, and a domain wall DW at the boundary between the two magnetic domains. The domain wall displacement layer 10 is a layer that can magnetically record information by changing its magnetic state, for example. The domain wall displacement layer 10 is also called an analog layer or a magnetic recording layer.

[0033] The domain wall displacement layer 10 has a first magnetization region A1, a second magnetization region A2, and a third magnetization region A3.

[0034] The first magnetization region A1 has a magnetization M Ａ１ The first magnetization region A1 is a region in which the orientation direction of the magnetization M is fixed in one direction. The magnetization being fixed means that the magnetization does not reverse during normal operation of the domain wall motion device 100 (when no external force exceeding the expected value is applied). The first magnetization region A1 is, for example, a region of the domain wall motion layer 10 that overlaps with the first magnetization fixed layer 40 when viewed from the z direction. The magnetization M of the first magnetization region A1 Ａ１ is, for example, the magnetization M of the first magnetization fixed layer 40 ４０ is fixed by

[0035] The second magnetization region A2 has magnetization M Ａ２ The magnetization M of the second magnetization region A2 is fixed in one direction. Ａ２ The orientation direction of the magnetization M of the first magnetization region A1 is Ａ１ The magnetization M of the second magnetization region A2 is different from the orientation direction of the Ａ２ The orientation direction of the magnetization M of the first magnetization region A1 is, for example, Ａ１ The second magnetization region A2 is, for example, a region of the domain wall displacement layer 10 that overlaps with the second magnetization fixed layer 50 when viewed from the z direction. The magnetization M of the second magnetization region A2 Ａ２ is, for example, the magnetization M of the second magnetization fixed layer 50 ５０ is fixed by

[0036] The third magnetization region A3 is a region other than the first magnetization region A1 and the second magnetization region A2 of the domain wall displacement layer 10. The third magnetization region A3 is, for example, a region sandwiched between the first magnetization region A1 and the second magnetization region A2 in the x direction.

[0037] The third magnetization region A3 is a region where the magnetization direction can change and the domain wall DW can move. The third magnetization region A3 is called a domain wall movable region. The third magnetization region A3 has a first magnetic domain A31 and a second magnetic domain A32. The first magnetic domain A31 and the second magnetic domain A32 have magnetization orientation directions opposite to each other. The boundary between the first magnetic domain A31 and the second magnetic domain A32 is the domain wall DW. The magnetization M of the first magnetic domain A31 Ａ３１ is, for example, the magnetization M of the first magnetization region A1 Ａ１ The magnetization M of the second magnetic domain A32 is oriented in the same direction as Ａ３２ is, for example, the magnetization M of the adjacent second magnetization region A2. Ａ２ In principle, the domain wall DW moves within the third magnetization region A3 and does not invade the first magnetization region A1 or the second magnetization region A2.

[0038] When the volume ratio between the first magnetic domain A31 and the second magnetic domain A32 in the third magnetization region A3 changes, the domain wall DW moves. The domain wall DW moves by passing a write current in the x direction of the third magnetization region A3. For example, when a write current (e.g., a current pulse) in the +x direction is applied to the third magnetization region A3, electrons flow in the -x direction, which is opposite to the current, and the domain wall DW moves in the -x direction. When a current flows from the first magnetic domain A31 to the second magnetic domain A32, electrons spin-polarized in the second magnetic domain A32 are attracted to the magnetization M of the first magnetic domain A31. Ａ３１ The magnetization of the first magnetic domain A31 is reversed. Ａ３１ By reversing the direction, the domain wall DW moves in the +x direction.

[0039] The domain wall displacement layer 10 is made of a magnetic material. The domain wall displacement layer 10 may be a ferromagnetic material, a ferrimagnetic material, or a combination of these with an antiferromagnetic material whose magnetic state can be changed by current. The domain wall displacement layer 10 preferably contains at least one element selected from the group consisting of Co, Ni, Fe, Pt, Pd, Gd, Tb, Mn, Ge, and Ga.

[0040] Examples of materials used for the domain wall displacement layer 10 include a laminated film of Co and Ni, a laminated film of Co and Pt, a laminated film of Co and Pd, and Co ｘ Fe １－ｘ Examples of the antiferromagnetic material include a laminated film of B (0≦x≦1) and the same material as the non-magnetic layer 20 described later, an MnGa-based material, a GdCo-based material, and a TbCo-based material. Ferrimagnetic materials such as MnGa-based materials, GdCo-based materials, and TbCo-based materials have small saturation magnetization, and the threshold current required to move the domain wall DW is small. Furthermore, a laminated film of Co and Ni, a laminated film of Co and Pt, and a laminated film of Co and Pd have large coercive force, and the moving speed of the domain wall DW is slow. Examples of the antiferromagnetic material include Mn ３ X (X is Sn, Ge, Ga, Pt, Ir, etc.), CuMnAs, Mn ２ Au or the like. The same material as the reference layer 30 described later can also be applied to the domain wall displacement layer 10. The domain wall displacement layer 10 can also be applied to two or more types of stacked films or materials.

[0041] The non-magnetic layer 20 is sandwiched in the z direction between the domain wall displacement layer 10 and the reference layer 30. The non-magnetic layer 20 inhibits magnetic coupling between the domain wall displacement layer 10 and the reference layer 30. The non-magnetic layer 20 is stacked on one surface of the reference layer 30.

[0042] The non-magnetic layer 20 is made of, for example, a non-magnetic insulator, a semiconductor, or a metal. The non-magnetic layer 20 is preferably made of, for example, a non-magnetic insulator. The non-magnetic insulator is, for example, Al. ２ O ３ , SiO ２ , MgO, MgAl ２ O ４ , and materials in which part of the Al, Si, and Mg in these are substituted with Zn, Be, Ga, Ti, etc. These materials have a large band gap and excellent insulating properties. The nonmagnetic insulator is, for example, an oxide containing Mg or Al. When the nonmagnetic layer 20 is made of a nonmagnetic insulator, the nonmagnetic layer 20 is a tunnel barrier layer. The nonmagnetic metal is, for example, Cu, Au, Ag, etc. The nonmagnetic semiconductor is, for example, Si, Ge, CuInSe ２ , CuGaSe ２ , Cu(In,Ga)Se ２ etc.

[0043] The thickness of the nonmagnetic layer 20 is, for example, 20 Å or more, and may be 25 Å or more. The thickness of each layer is the average value of the height in the z direction of the layer measured at five different points in the x direction.

[0044] The reference layer 30 sandwiches the nonmagnetic layer 20 together with the domain wall displacement layer 10. The reference layer 30 is located so that at least a portion thereof overlaps with the third magnetization region A3 in the z direction. The reference layer 30 is, for example, closer to the substrate Sub than the domain wall displacement layer 10.

[0045] 5 and 6 , the side surface of the reference layer 30 is inclined with respect to the z direction. The width in the y direction and the length in the x direction of the reference layer 30 increase with increasing distance from the nonmagnetic layer 20 (closer to the substrate Sub). The width W30 in the y direction of the second surface of the reference layer 30 farther from the nonmagnetic layer 20 is longer than the width in the y direction of the first surface of the reference layer 30 on the nonmagnetic layer 20 side.

[0046] The width W30 satisfies 3×W10<W30<L10. As described above, L10 is the length of the domain wall displacement layer 10 in the x direction, and W10 is the width of the domain wall displacement layer 10 in the y direction. When this relationship is satisfied, the influence of the leakage magnetic field from the reference layer 30 on the domain wall displacement layer 10 can be reduced.

[0047] The reference layer 30 includes, for example, a first ferromagnetic layer 31 , an antiferromagnetic coupling layer 33 , and a second ferromagnetic layer 32 .

[0048] The first ferromagnetic layer 31 and the second ferromagnetic layer 32 include a ferromagnetic material. The first ferromagnetic layer 31 and the second ferromagnetic layer 32 include, for example, a metal selected from the group consisting of Cr, Mn, Co, Fe, and Ni, an alloy containing one or more of these metals, or an alloy containing one or more of these metals and at least one of B, C, and N. The first ferromagnetic layer 31 and the second ferromagnetic layer 32 are, for example, Co—Fe, Co—Fe—B, or Ni—Fe. The first ferromagnetic layer 31 and the second ferromagnetic layer 32 may also include a stacked film of Co and Ni, a stacked film of Co and Pt, or a stacked film of Co and Pd.

[0049] The first ferromagnetic layer 31 and the second ferromagnetic layer 32 may be, for example, a Heusler alloy. The Heusler alloy is half-metallic and has high spin polarization. The Heusler alloy is an XYZ or X２ It is an intermetallic compound having a chemical composition of YZ, where X is a transition metal element or a noble metal element of the Co, Fe, Ni, or Cu group on the periodic table, Y is a transition metal element or an element species of X of the Mn, V, Cr, or Ti group, and Z is a typical element of groups III to V. ２ FeSi, Co ２ FeGe, Co ２ FeGa, Co ２ MnSi, Co ２ Mn １－ａ Fe ａ Al ｂ Si １－ｂ , Co ２ FeGe １－ｃ Ga ｃ etc.

[0050] The magnetization M of the first ferromagnetic layer 31 ３１ is the magnetization M of the second ferromagnetic layer 32 ３２ Therefore, the magnetization M ３１ is the magnetization M ３２ is oriented in the opposite direction.

[0051] For example, the product of the thickness and saturation magnetization of the first ferromagnetic layer 31 may be equal to the product of the thickness and saturation magnetization of the second ferromagnetic layer 32. When this relationship is satisfied, the magnitudes of the magnetic fields from the first ferromagnetic layer 31 and the second ferromagnetic layer 32 are substantially equal due to a mutually canceling relationship, thereby reducing the leakage magnetic field from the reference layer 30. Furthermore, for example, the product of the thickness and saturation magnetization of the first ferromagnetic layer 31 may be greater than the product of the thickness and saturation magnetization of the second ferromagnetic layer 32. When this relationship is satisfied, the magnetic balance between the first ferromagnetic layer 31 and the second ferromagnetic layer 32 is shifted, making it easier to control the magnetization orientation direction of the reference layer 30. Even if the magnetic balance between the first ferromagnetic layer 31 and the second ferromagnetic layer 32 is shifted, the leakage magnetic field from the reference layer 30 can be sufficiently reduced by satisfying the relationship 3×W10<W30<L10.

[0052] The magnetization M of the first ferromagnetic layer 31 ３１ and the magnetization M of the second ferromagnetic layer 32 ３２The magnetization M of the first ferromagnetic layer 31 is more difficult to reverse than the magnetization of the third magnetization region A3 of the domain wall displacement layer 10 due to antiferromagnetic coupling. ３１ and the magnetization M of the second ferromagnetic layer 32 ３２ The reference layer 30 is fixed and does not change its direction when an external force is applied strong enough to reverse the magnetization of the third magnetization region A3. The reference layer 30 is also sometimes called a fixed layer.

[0053] 6, in the y-z cross section, the first ferromagnetic layer 31 can be divided into a first central region R1 and first outer peripheral regions R2 and R3. The first central region R1 is a region that overlaps with the domain wall displacement layer 10 in the z direction. The first outer peripheral regions R2 and R3 are regions that do not overlap with the domain wall displacement layer 10 in the z direction. The first central region R1 is sandwiched between the first outer peripheral region R2 and the first outer peripheral region R3 in the y direction.

[0054] 6, in the y-z cross section, the second ferromagnetic layer 32 can be divided into a second central region R4 and second outer peripheral regions R5 and R6. The second central region R4 is a region that overlaps with the domain wall displacement layer 10 in the z direction. The second outer peripheral regions R5 and R6 are regions that do not overlap with the domain wall displacement layer 10 in the z direction. The second central region R4 is sandwiched between the second outer peripheral regions R5 and R6 in the y direction.

[0055] The first central region R1 and the second central region R4 face each other and are magnetically coupled. In contrast, the first outer peripheral region R2 and the second outer peripheral region R5 are regions where magnetic balance is disrupted and they can become sources of leakage magnetic fields. Similarly, the first outer peripheral region R3 and the second outer peripheral region R6 are regions where magnetic balance is disrupted and they can become sources of leakage magnetic fields. Therefore, it is preferable that the magnetization of the first outer peripheral regions R2 and R3 is smaller than that of the first central region R1, and it is preferable that the magnetization of the second outer peripheral regions R5 and R6 is smaller than that of the second central region R4. Magnetization is the magnetic moment per unit volume.

[0056] The magnetization of the first outer peripheral regions R2 and R3 can be made smaller than the magnetization of the first central region R1, for example, by implanting a nonmagnetic element. The magnetization of the second outer peripheral regions R5 and R6 can be made smaller than the magnetization of the second central region R4, for example, by implanting a nonmagnetic element. The first outer peripheral regions R2 and R3 preferably contain a nonmagnetic element. The second outer peripheral regions R5 and R6 preferably contain a nonmagnetic element. The nonmagnetic element is, for example, one or more elements selected from the group consisting of Ta, Ru, Pt, Pd, Ti, Cu, Ir, Mo, W, and Au.

[0057] The first outer peripheral regions R2, R3 and the second outer peripheral regions R5, R6 may be oxidized or nitrided to reduce the magnetization of these regions. The first outer peripheral regions R2, R3 preferably contain an oxide or nitride. The second outer peripheral regions R5, R6 preferably contain an oxide or nitride. The oxide or nitride is, for example, an oxide or nitride of the ferromagnetic material constituting the first ferromagnetic layer 31 or the second ferromagnetic layer 32.

[0058] Here, the case has been exemplified in which the magnetization of the entire first outer peripheral regions R2, R3 is smaller than that of the first central region R1, and the magnetization of the entire second outer peripheral regions R5, R6 is smaller than that of the second central region R4, but the portion with small magnetization may be a portion of the first outer peripheral regions R2, R3 or the second outer peripheral regions R5, R6. For example, the portion with small magnetization may be a portion of the first outer peripheral region R2 that does not face the second outer peripheral region R5, or a portion of the first outer peripheral region R3 that does not face the second outer peripheral region R6.

[0059] For example, the concentration of the nonmagnetic element contained in the first outer peripheral regions R2, R3 may increase with increasing distance from the first central region R1. For example, the concentration of the oxide or nitride contained in the first outer peripheral regions R2, R3 may increase with increasing distance from the first central region R1. Similarly, the concentration of the nonmagnetic element contained in the second outer peripheral regions R5, R6 may increase with increasing distance from the second central region R4. For example, the concentration of the oxide or nitride contained in the second outer peripheral regions R5, R6 may increase with increasing distance from the second central region R4.

[0060] The antiferromagnetic coupling layer 33 is sandwiched between the first ferromagnetic layer 31 and the second ferromagnetic layer 32 in the z direction. The antiferromagnetic coupling layer 33 is a non-magnetic metal layer. The antiferromagnetic coupling layer 33 is made of, for example, Ru, Ir, or Rh. The thickness of the antiferromagnetic coupling layer 33 is, for example, a thickness that causes antiferromagnetic coupling between the first ferromagnetic layer 31 and the second ferromagnetic layer 32 through RKKY interaction.

[0061] The reference layer 30 and the non-magnetic layer 20 are each longer than the third magnetization region A3 in the x direction, for example. The portion where the reference layer 30 and the third magnetization region A3 face each other with the non-magnetic layer 20 sandwiched therebetween is responsible for the resistance change of the domain wall motion element 100. When the length of the reference layer 30 and the non-magnetic layer 20 in the x direction is longer than the length of the third magnetization region A3 in the x direction, the resistance change of the domain wall motion element 100 becomes gentler, making it easier to divide the resistance change range of the domain wall motion element 100 into more values.

[0062] The reference layer 30 is, for example, longer in the x direction than the domain wall motion layer 10. When the reference layer 30 overlaps the entire domain wall motion layer 10 as viewed in the z direction, the heat dissipation of the domain wall motion layer 10 is improved. As a result, the stability of the magnetization of the first magnetization region A1 and the magnetization of the second magnetization region A2 is improved, and the reliability of the data of the domain wall motion element 100 is improved.

[0063] The first magnetization pinned layer 40 is connected to the domain wall displacement layer 10. The first magnetization pinned layer 40 is connected to the first magnetization region A1. The first magnetization pinned layer 40 is connected to the magnetization M of the first magnetization region A1. Ａ１ There is no particular limitation on the shape of the first magnetization fixed layer 40 in a plan view. The shape of the first magnetization fixed layer 40 in a plan view may be, for example, rectangular as shown in FIG. 4 or circular.

[0064] The second magnetization pinned layer 50 is connected to the domain wall displacement layer 10 at a position spaced apart from the first magnetization pinned layer 40 in the x direction. The second magnetization pinned layer 50 is connected to the second magnetization region A2. The second magnetization pinned layer Ａ２ The shape of the second magnetization fixed layer 50 in plan view may be, for example, rectangular as shown in FIG. 5 or circular.

[0065] The film thickness of the first magnetization pinned layer 40 and the film thickness of the second magnetization pinned layer 50 may be different. For example, the film thickness of the first magnetization pinned layer 40 is thicker than the film thickness of the second magnetization pinned layer 50. This difference in film thickness is utilized to create a difference in coercivity between the first magnetization pinned layer 40 and the second magnetization pinned layer 50.

[0066] Here, the positional relationship between the first magnetization pinned layer 40 and the second magnetization pinned layer 50 is not limited to the example shown in Fig. 5. The positional relationship between the first magnetization pinned layer 40 and the second magnetization pinned layer 50 may be reversed, and the first magnetization pinned layer 40 may be located in the +x direction of the second magnetization pinned layer 50.

[0067] The first electrode E1 is connected to the first magnetization fixed layer 40. The first electrode E1 is, for example, a write electrode electrically connected to a write wiring WL. The second electrode E2 is connected to the second magnetization fixed layer 50. The second electrode E2 is, for example, a common electrode electrically connected to a common wiring CL. The third electrode E3 is connected to the reference layer 30. The third electrode E3 is, for example, a read electrode electrically connected to a read wiring RL.

[0068] The domain wall motion element 100 may have layers other than those described above. For example, a base layer may be provided below the reference layer 30, or a cap layer may be provided above the domain wall motion layer 10.

[0069] The magnetization direction of each layer of the domain wall motion element 100 can be confirmed by, for example, measuring a magnetization curve. The magnetization curve can be measured using, for example, MOKE (Magneto-Optical Kerr Effect). Measurement using MOKE is a measurement method that uses the magneto-optical effect (magnetic Kerr effect) in which linearly polarized light is incident on a measurement object and the polarization direction rotates.

[0070] The domain wall motion element 100 is formed by a lamination process of each layer and a processing process of processing a part of each layer into a predetermined shape.

[0071] In the lamination process, each layer constituting the domain wall motion element 100 is formed. The layers can be formed by sputtering, chemical vapor deposition (CVD), electron beam evaporation (EB evaporation), atomic laser deposition, or the like.

[0072] In the processing step, a portion of the stack formed in the stacking step is processed. The stack can be processed using photolithography, etching (e.g., Ar etching), or the like. The length L10 and width W10 of the domain wall displacement layer 10 and the width W30 of the reference layer 30 can be freely designed by changing the size of the mask, the irradiation angle of the etching ion beam, and the like.

[0073] Next, the operation of writing a signal to the magnetic array MA and the operation of reading a signal from the magnetic array MA will be described.

[0074] First, the operation of writing a signal to the magnetic array MA will be described. The write operation is performed by, for example, having a processor execute an operation program stored in the control unit 6.

[0075] First, the control device 3 selects the domain wall motion element 100 to which a pulse is to be applied in accordance with the operation program. When the magnetic array MA is used as a magnetic memory, the domain wall motion element 100 to which the pulse is applied is an element that stores data. When the magnetic array MA is used as a neural network, the domain wall motion element 100 to which the pulse is applied is an element that changes the weight in accordance with learning.

[0076] The control unit 6 controls which of the multiple domain wall motion elements 100 to apply a pulse to. The control unit 6 turns on the first switch SW1 and the second switch SW2 connected to the domain wall motion element 100 to which the pulse is to be applied, and turns off the third switch SW3. The control unit 6 also turns off at least one of the first switch SW1 and the second switch SW2 connected to the domain wall motion element 100 to which the pulse is not to be applied.

[0077] Then, the control device 3 outputs a write pulse to the domain wall motion element 100 in accordance with the operation program. The write pulse is applied between the first magnetization pinned layer 40 and the second magnetization pinned layer 50 along the domain wall motion layer 10 of the domain wall motion element 100. The write pulse may be a square wave, a spike wave, or a wave with another waveform. By changing the number of write pulses, the magnitude, etc., the position of the domain wall DW changes, and a signal is written to a specific domain wall motion element 100.

[0078] Next, the operation of reading a signal from the magnetic array MA will be described. The reading operation is performed by, for example, having a processor execute an operation program stored in the control unit 6.

[0079] First, the control device 3 selects the domain wall motion element 100 to which a read pulse is applied in accordance with the operation program. When the magnetic array MA is used as a magnetic memory, the domain wall motion element 100 to which the read pulse is applied is an element that reads data. When the magnetic array MA is used as a neural network, the application of a read pulse to a specific domain wall motion element 100 corresponds to a multiplication operation of the input and the weight. In other words, when the magnetic array MA is used as a neural network, the read operation is an identification operation of the neural network.

[0080] The control unit 6 controls which of the multiple domain wall motion elements 100 to apply a pulse to. The control unit 6 turns on the third switch SW3 and the second switch SW2 connected to the domain wall motion element 100 to which the pulse is to be applied, and turns off the first switch SW1. The control unit 6 also turns off at least one of the third switch SW3 and the second switch SW2 connected to the domain wall motion element 100 to which the pulse is not to be applied.

[0081] Next, the control device 3 applies a read pulse to a predetermined domain wall motion element 100 in accordance with the operation program. The read pulse is applied, for example, between the third electrode E3 and the second magnetization fixed layer 50. The voltage of the read pulse is a voltage that obtains a current density less than the critical current density required to move the domain wall DW of the domain wall motion layer 10. In other words, the read pulse does not move the domain wall DW.

[0082] The resistance detection device 4 detects the resistance value of the domain wall motion element 100 to which the read pulse has been applied. The output unit 5 outputs the calculation result to the outside, for example. By this procedure, a signal can be read out from a specific domain wall motion element 100.

[0083] In the domain wall motion element 100 according to this embodiment, the reference layer 30 has an inclined side surface, and satisfies the relationship 3×W10<W30<L. Therefore, the domain wall motion element 100 according to this embodiment can reduce the leakage magnetic field applied from the reference layer 30 to the domain wall motion layer 10.

[0084] Fig. 7 shows the results of a simulation of a leakage magnetic field applied to the domain wall displacement layer 10. Fig. 8 is a schematic diagram of a model used in the simulation of the leakage magnetic field.

[0085] The first model is the result of a simulation model of a domain wall motion element in which the width W10 of the domain wall motion layer 10 and the width W30 of the reference layer 30 are the same. In the domain wall motion element in the first model, the domain wall motion layer 10 and the reference layer 30 overlap when viewed from the z direction. The second model is the result of a simulation model of a domain wall motion element in which the width W30 of the reference layer 30 is larger than the width W10 of the domain wall motion layer 10 and the side surface of the reference layer 30 is not inclined. The third model is the result of a simulation model of a domain wall motion element in which the width W30 of the reference layer 30 is larger than the width W10 of the domain wall motion layer 10 and the side surface of the reference layer 30 is inclined. In the second and third models, W30 = 2 × W10.

[0086] The second model has a smaller fringing magnetic field than the first and third models. There are at least two possible reasons why the fringing magnetic field of the second model is smaller. The first reason is that the width W30 of the reference layer 30 is sufficiently larger than the width W10 of the domain wall displacement layer 10. When this relationship is satisfied, the fringing magnetic field from the side surface of the reference layer 30 can be prevented from reaching the domain wall displacement layer 10. The second reason is that the volume difference between the first ferromagnetic layer 31 and the second ferromagnetic layer 32 is small at any position in the y direction. When this relationship is satisfied, the magnitudes of the fringing magnetic field generated from the first ferromagnetic layer 31 and the fringing magnetic field generated from the second ferromagnetic layer 32 become approximately equal, and the fringing magnetic fields from each layer cancel each other out, thereby suppressing the fringing magnetic field from the entire reference layer 30.

[0087] From the above, the shape of the second model is ideal for reducing the leakage magnetic field from the reference layer 30. On the other hand, the second model requires a mask for processing the outer shape of the domain wall displacement layer 10 and a mask for processing the outer shape of the reference layer 30, which complicates the manufacturing process.

[0088] When the domain wall displacement layer 10 and the reference layer 30 are formed using a single mask, the shape of the third model is obtained. In the third model, the side surface of the reference layer 30 is inclined, resulting in a significant volume difference between the first ferromagnetic layer 31 and the second ferromagnetic layer 32 at any position in the y direction. On the other hand, when the reference layer 30 is formed so that 3×W10<W30 is satisfied, the side surface is gently inclined. With a gentle side surface inclination, the volume difference between the first ferromagnetic layer 31 and the second ferromagnetic layer 32 at any position in the y direction is sufficiently small. As a result, a reference layer 30 that satisfies 3×W10<W30 can sufficiently reduce the leakage magnetic field to the domain wall displacement layer 10 compared to the first model. Furthermore, when 3×W10<W30 is satisfied, a portion where the first ferromagnetic layer 31 and the second ferromagnetic layer 32 do not face each other and a leakage magnetic field occurs is formed at a position away from the domain wall displacement layer 10. As a result, the influence of the leakage magnetic field from this portion on the domain wall displacement layer 10 is reduced.

[0089] The domain wall motion element 100 according to the first embodiment can be used in, for example, a magnetic memory or a neuromorphic device.

[0090] In the case of a magnetic memory, each of the domain wall motion elements 100 functions as an element for storing data. The resistance of the domain wall motion element 100 changes depending on the position of the domain wall DW of the domain wall motion element 100, and this resistance value is used to store data.

[0091] In the case of a neuromorphic device, each of the domain wall motion elements 100 functions as a product calculation element. The resistance of the domain wall motion element 100 changes depending on the position of the domain wall DW of the domain wall motion element, and this resistance value represents the weight.

[0092] A neuromorphic device is a device that artificially mimics the relationship between neurons and synapses in the human brain. Neuromorphic devices are capable of performing neural network calculations.

[0093] 9 is a schematic diagram of the neural network NN. The neural network NN has an input layer L ｉｎ and the middle layer L ｍ and the output layer L ｏｕｔ In FIG. ｍ presents an example of three layers, but the middle layer L ｍ The number of input layers L ｉｎ and the middle layer L ｍ and the output layer L ｏｕｔ Each of the input layers L has a plurality of chips C, each of which corresponds to a neuron in the brain. ｉｎ and the middle layer L ｍ and the output layer L ｏｕｔ The chips C and the number of transmission means shown in FIG. 9 are merely examples.

[0094] The neural network NN increases the rate of correct answers to questions as its transmission means (synapses) learn. Learning is the process of finding knowledge that may be useful in the future from information. The neural network NN learns by operating while changing the weight of the transmission means. The transmission means performs a multiplication operation to apply a weight to the input signal, and a sum operation to add the result of the multiplication operation. In other words, the transmission means performs a product-sum operation. The domain wall motion device 100 according to this embodiment is responsible for this multiplication operation.

[0095] 10 is a block diagram showing a system 300 including the neuromorphic device 200 according to the first embodiment. The system 300 includes a plurality of sensors 201, the neuromorphic device 200, and a communication unit 202.

[0096] Any sensor can be used for each of the multiple sensors 201 depending on the application. For example, a temperature sensor, a humidity sensor, a speed sensor, a pressure sensor, an acceleration sensor, etc. can be used as the multiple sensors 201. The signals from these sensors are input to, for example, the input layer L ｉｎ corresponds to the signal input to

[0097] The neuromorphic device 200 has, for example, multiple accumulation regions 1. Each accumulation region 1 performs a product-sum operation. Each accumulation region 1 performs operations from each layer of the neural network NN to the next layer. Each accumulation region 1 may have a separate control device 3, or may share the control device 3.

[0098] The conductance (or resistance) of the domain wall motion element 100 changes depending on the position of the domain wall DW. The conductance (or resistance) of the domain wall motion element 100 corresponds to the weight of the transmission means in the neural network NN. The conductance (or resistance) of the domain wall motion element 100 changes linearly with respect to the input. For example, if information (e.g., temperature) from a specific sensor 201 among the multiple sensors 201 is important, the conductance (weight) of the domain wall motion element 100 responsible for transmitting the signal from that sensor 201 is increased when the neuromorphic device 200 learns.

[0099] The domain wall motion element 100 functions as a product calculation element because it outputs a signal that is the product of the input voltage and the conductance (or resistance) of the domain wall motion element 100 itself. The magnetic array MA functions as a product-sum calculator because it combines the outputs from multiple domain wall motion elements 100. The product-sum calculation by the multiple domain wall motion elements 100 is controlled by the control device 3.

[0100] The neuromorphic device 200 performs learning and inference. The conductance (corresponding to the weight of the transmission means) of the domain wall motion element 100 is adjusted during learning. Inference is performed using the set conductance (corresponding to the weight of the transmission means) of the domain wall motion element 100.

[0101] The neuromorphic device 200 used in the system 300 may be capable of both learning and inference, or may be capable of only inference. When only inference is performed, learning tailored to the task is performed in advance, and weights tailored to the task are installed in the domain wall motion elements 100 of the neuromorphic device 200. For example, the conductance of each domain wall motion element 100 is adjusted so as to correspond to the weights of the transmission means determined in the advance learning. If the neuromorphic device 200 is capable of only inference, the computational load on the edge device can be reduced.

[0102] The communication unit 202 outputs the calculation results of the neuromorphic device 200 to the outside. For example, an inference result for a predetermined task obtained by the neuromorphic device 200 is input to the communication unit 202, and the communication unit 202 outputs the information to the outside. The communication unit 202 may be wired or wireless.

[0103] The domain wall motion device 100 according to this embodiment is less affected by the leakage magnetic field and the motion of the domain wall DW is less likely to be disturbed, so the reliability of the system 300 is high.

[0104] 11 is a yz cross-sectional view of a domain wall motion element 101 according to a second embodiment. The domain wall motion element 101 can be replaced with the domain wall motion element 100. The domain wall motion element 101 has a reference layer 30A that is different from the reference layer 30 of the domain wall motion element 100. In the second embodiment, the same components as those in the first embodiment are denoted by the same reference symbols, and descriptions thereof will be omitted.

[0105] The reference layer 30A includes a first ferromagnetic layer 31A, a second ferromagnetic layer 32A, and an antiferromagnetic coupling layer 33. The first ferromagnetic layer 31A and the second ferromagnetic layer 32A are similar to the first ferromagnetic layer 31 and the second ferromagnetic layer 32 except for their thicknesses. The thickness of the first ferromagnetic layer 31A is thinner than the thickness of the second ferromagnetic layer 32A. The magnitude of saturation magnetization of the first ferromagnetic layer 31A is greater than the magnitude of saturation magnetization of the second ferromagnetic layer 32A.

[0106] The product of the thickness and saturation magnetization of the first ferromagnetic layer 31A may be equal to the product of the thickness and saturation magnetization of the second ferromagnetic layer 32A, or may be greater than the product of the thickness and saturation magnetization of the second ferromagnetic layer 32A.

[0107] 11, the film thicknesses of the first ferromagnetic layer 31A and the second ferromagnetic layer 32A may be different. The domain wall motion element 101 according to the second embodiment has the same effects as the domain wall motion element 100 according to the first embodiment.

[0108] 12 is a yz cross-sectional view of a domain wall motion element 102 according to a third embodiment. The domain wall motion element 102 can be replaced with the domain wall motion element 100. The domain wall motion element 102 has a reference layer 30B that is different from the reference layer 30 of the domain wall motion element 100. In the third embodiment, the same components as those in the first embodiment are denoted by the same reference symbols, and descriptions thereof will be omitted.

[0109] The reference layer 30B has a first inclined surface 31s and a second inclined surface 32s. The intersection of the first inclined surface 31s and the second inclined surface 32s is curved. The first inclined surface 31s is located farther from the nonmagnetic layer 20 than the second inclined surface 32s. The first inclined surface 31s is, for example, a side surface of the first ferromagnetic layer 31B. The second inclined surface 32s is, for example, a side surface of the second ferromagnetic layer 32B. The first ferromagnetic layer 31B and the second ferromagnetic layer 32B are similar to the first ferromagnetic layer 31 and the second ferromagnetic layer 32.

[0110] The inclination angle of the second inclined surface 32s with respect to the stacking surface is greater than the inclination angle of the first inclined surface 31s with respect to the stacking surface. The stacking surface is a surface parallel to the xy plane. The inclination angles of the first inclined surface 31s and the second inclined surface 32s do not have to be constant. In this case, the average value of the inclination angles measured at three different points in the y direction is treated as the inclination angle.

[0111] If the tilt angle of the second sloping surface 32s is larger than the tilt angle of the first sloping surface 31s, the volume difference between the first ferromagnetic layer 31B and the second ferromagnetic layer 32B can be made sufficiently small at a position close to the domain wall motion layer 10. As a result, the leakage magnetic field generated from the reference layer 30B is reduced, and the influence of the leakage magnetic field on the domain wall motion layer 10 can be reduced. The domain wall motion element 102 according to the third embodiment achieves the same effects as the domain wall motion element 100 according to the first embodiment.

[0112] 13 is a yz cross-sectional view of a domain wall motion element 103 according to a fourth embodiment. The domain wall motion element 103 can be replaced with the domain wall motion element 100. The domain wall motion element 103 has a reference layer 30C that is different from the reference layer 30 of the domain wall motion element 100. In the fourth embodiment, the same components as those in the first embodiment are denoted by the same reference symbols, and descriptions thereof will be omitted.

[0113] The reference layer 30C includes a first ferromagnetic layer 31C, a second ferromagnetic layer 32C, and an antiferromagnetic coupling layer 33.

[0114] The first ferromagnetic layer 31C includes a plurality of ferromagnetic layers 311 and 312 and an intermediate layer 313. The ferromagnetic layers 311 and 312 are magnetically coupled with each other with the intermediate layer 313 sandwiched therebetween.

[0115] 13 shows an example in which the ferromagnetic layers 311 and 312 are ferromagnetically coupled, but the ferromagnetic layers 311 and 312 may be antiferromagnetically coupled. Also, while an example in which there are two ferromagnetic layers is shown in FIG. 13, the number of ferromagnetic layers is not particularly limited.

[0116] The ferromagnetic layers 311 and 312 may be made of the same material as the first ferromagnetic layer 31. The intermediate layer 313 may be made of the same material as the antiferromagnetic coupling layer 33, for example.

[0117] The second ferromagnetic layer 32C includes a plurality of ferromagnetic layers 321 and 322 and an intermediate layer 323. The ferromagnetic layers 321 and 322 are magnetically coupled with each other with the intermediate layer 323 sandwiched therebetween.

[0118] 13 shows an example in which the ferromagnetic layers 321 and 322 are ferromagnetically coupled, but the ferromagnetic layers 321 and 322 may be antiferromagnetically coupled. Also, while the example in FIG. 13 shows two ferromagnetic layers, the number of ferromagnetic layers is not particularly limited.

[0119] The ferromagnetic layer 321 and the ferromagnetic layer 322 can be made of the same material as the first ferromagnetic layer 31. Of the multiple ferromagnetic layers 321 and 322, the ferromagnetic layer 322 closest to the non-magnetic layer 20 preferably contains CoFeB. This configuration increases the MR ratio of the domain wall motion element 100. The intermediate layer 313 can be made of, for example, the same material as the antiferromagnetic coupling layer 33.

[0120] When the reference layer 30 is made up of multiple layers, the thickness of the reference layer 30 can be increased, and the inclination of the side surface of the reference layer 30 can be made gentler. As a result, it becomes easier to realize a configuration in which the width W30 of the reference layer 30 is larger than the width W10 of the domain wall motion layer 10. The domain wall motion element 103 according to the fourth embodiment achieves the same effects as the domain wall motion element 100 according to the first embodiment.

[0121] 14 is a plan view of a domain wall motion element 104 according to a fifth embodiment. The domain wall motion element 104 can be replaced with the domain wall motion element 100. The domain wall motion element 104 has a reference layer 30D whose shape differs from that of the reference layer 30 of the domain wall motion element 100. In the fifth embodiment, the same components as those in the first embodiment are denoted by the same reference symbols, and a description thereof will be omitted.

[0122] The reference layer 30D is similar to the reference layer 30 except for its shape when viewed from the z direction. The reference layer 30D does not have a constant width in the y direction when viewed from the z direction. The width W30E in the y direction of the reference layer 30D at a position where it overlaps with the first magnetization fixed layer 40 in the z direction is wider than the width W30C in the y direction of the reference layer 30D at a position where it does not overlap with the first magnetization fixed layer 40 in the z direction. This shape can be achieved by utilizing the difference in etching rate between the first magnetization fixed layer 40 and the insulating layer 90 on the domain wall displacement layer 10.

[0123] The magnetization at the position overlapping with the first magnetization fixed layer 40 is more strongly fixed than the magnetization at the position not overlapping. In other words, the magnetization at the position not overlapping with the first magnetization fixed layer 40 is more susceptible to the influence of the leakage magnetic field. If the width W30E is greater than the width W30C, the leakage magnetic field generated in the portion where the magnetization is relatively weakly fixed can be reduced. The domain wall motion element 104 according to the fifth embodiment has the same effect as the domain wall motion element 100 according to the first embodiment.

[0124] 15 is a plan view of a domain wall motion element 105 according to a sixth embodiment. The domain wall motion element 105 can be replaced with the domain wall motion element 100. The domain wall motion element 105 has a reference layer 30E whose shape differs from that of the reference layer 30 of the domain wall motion element 100. In the sixth embodiment, the same components as those in the first embodiment are denoted by the same reference symbols, and a description thereof will be omitted.

[0125] The reference layer 30E is similar to the reference layer 30 except for its shape when viewed from the z direction. The reference layer 30E does not have a constant width in the y direction when viewed from the z direction. The width W30E in the y direction of the reference layer 30E at a position where it overlaps with the first magnetization fixed layer 40 in the z direction is narrower than the width W30C in the y direction of the reference layer 30 at a position where it does not overlap with the first magnetization fixed layer 40 in the z direction. This shape can be achieved by utilizing the difference in etching rate between the first magnetization fixed layer 40 and the insulating layer 90 on the domain wall displacement layer 10.

[0126] Most of the heat generated in the domain wall motion layer 10 is dissipated via the first magnetization fixed layer 40 or the second magnetization fixed layer 50. Therefore, heat generated at a position overlapping with the first magnetization fixed layer 40 is more difficult to dissipate than heat generated at a position not overlapping. If the width W30C is greater than the width W30E, heat can be efficiently dissipated even from areas where it is relatively difficult to dissipate heat. The domain wall motion element 105 according to the sixth embodiment has the same effects as the domain wall motion element 100 according to the first embodiment.

[0127] Although the preferred embodiments of the present disclosure have been described in detail above, the present disclosure is not limited to these embodiments. For example, the characteristic configurations of the above embodiments may be combined with each other.

[0128] REFERENCE SIGNS LIST 1 Accumulation region 2 Peripheral region 3 Control device 4 Resistance detection device 5 Output unit 6 Control unit 7 Power supply 10 Domain wall motion layer 20 Non-magnetic layer 30, 30A, 30B, 30C, 30D, 30E Reference layer 31, 31A, 31B, 31C First ferromagnetic layer 31s First inclined surface 32, 32A, 32B, 32C Second ferromagnetic layer 32s Second inclined surface 33 Antiferromagnetic coupling layer 40 First magnetization fixed layer 50 Second magnetization fixed layer 90 Insulating layer 100, 101, 102, 103, 104, 105 Domain wall motion element 200 Neuromorphic device 201 Sensor 202 Communication unit 300 System 311, 312, 321, 322 Ferromagnetic layer 313, 323 Intermediate layer DW Domain wall E1 First electrode E2 Second electrode E3 Third electrode R1 First central region R2, R3 First outer peripheral region R4 Second central region R5, R6 Second outer peripheral region

Claims

1. A domain wall motion element comprising: a reference layer, a non-magnetic layer, and a domain wall motion layer, wherein the non-magnetic layer is stacked on the reference layer and sandwiched between the reference layer and the domain wall motion layer in the stacking direction, wherein the reference layer comprises a first ferromagnetic layer, an antiferromagnetic coupling layer, and a second ferromagnetic layer, wherein the antiferromagnetic coupling layer is sandwiched between the first ferromagnetic layer and the second ferromagnetic layer in the stacking direction, and wherein a side surface of the reference layer is inclined with respect to the stacking direction, wherein a first direction is a longitudinal direction of the domain wall motion layer, a second direction is a direction perpendicular to the stacking direction and the first direction, a length of the domain wall motion layer in the first direction is L10, a width in the second direction of a first surface of the domain wall motion layer remote from the non-magnetic layer is W10, and a width in the second direction of a second surface of the reference layer remote from the non-magnetic layer is W30, the relationship 3×W10<W30<L10 is satisfied.

2. The domain wall motion element described in claim 1, wherein the first ferromagnetic layer has, in a cross section perpendicular to the first direction, a first central region positioned to overlap with the domain wall motion layer in the stacking direction, and a first outer peripheral region positioned not to overlap with the domain wall motion layer in the stacking direction; the second ferromagnetic layer has, in a cross section perpendicular to the first direction, a second central region positioned to overlap with the domain wall motion layer in the stacking direction, and a second outer peripheral region positioned not to overlap with the domain wall motion layer in the stacking direction; the magnetization of the first outer peripheral region is smaller than the magnetization of the first central region; and the magnetization of the second outer peripheral region is smaller than the magnetization of the second central region.

3. The domain wall motion element according to claim 2, wherein the first outer peripheral region or the second outer peripheral region contains a non-magnetic element.

4. The domain wall motion element according to claim 3, wherein the non-magnetic element is one or more elements selected from the group consisting of Ta, Ru, Pt, Pd, Ti, Cu, Ir, Mo, W, and Au.

5. The domain wall motion element according to claim 2, wherein the first outer peripheral region or the second outer peripheral region includes an oxide or a nitride.

6. The domain wall motion element according to claim 1, wherein the film thickness of the first ferromagnetic layer is thinner than the film thickness of the second ferromagnetic layer, and the magnitude of the saturation magnetization of the first ferromagnetic layer is greater than the magnitude of the saturation magnetization of the second ferromagnetic layer.

7. The domain wall motion element according to claim 1, wherein a side surface of the reference layer has a first inclined surface and a second inclined surface, the first inclined surface is located farther from the nonmagnetic layer than the second inclined surface, and the inclination angle of the second inclined surface relative to the stacking surface is larger than the inclination angle of the first inclined surface relative to the stacking surface.

8. The domain wall motion element according to claim 1, wherein the product of the film thickness of the first ferromagnetic layer and the saturation magnetization of the first ferromagnetic layer is greater than the product of the film thickness of the second ferromagnetic layer and the saturation magnetization of the second ferromagnetic layer.

9. The domain wall motion element according to claim 1, wherein at least one of the first ferromagnetic layer and the second ferromagnetic layer includes a plurality of ferromagnetic layers magnetically coupled to each other.

10. The domain wall motion element according to claim 1, wherein the second ferromagnetic layer has a plurality of ferromagnetic layers, and of the plurality of ferromagnetic layers, the ferromagnetic layer closest to the nonmagnetic layer contains CoFeB.

11. The domain wall motion element according to claim 1, further comprising a first magnetization pinned layer and a second magnetization pinned layer, wherein the first magnetization pinned layer is connected to the domain wall motion layer, and the second magnetization pinned layer is connected to the domain wall motion layer at a position different in the first direction from the first magnetization pinned layer, and the width in the second direction of the reference layer at a position overlapping with the first magnetization pinned layer in the stacking direction is narrower than the width in the second direction of the reference layer at a position not overlapping with the first magnetization pinned layer in the stacking direction.

12. The domain wall motion element according to claim 1, further comprising a first magnetization pinned layer and a second magnetization pinned layer, wherein the first magnetization pinned layer is connected to the domain wall motion layer, and the second magnetization pinned layer is connected to the domain wall motion layer at a position different in the first direction from the first magnetization pinned layer, and the width in the second direction of the reference layer at a position overlapping with the first magnetization pinned layer in the stacking direction is wider than the width in the second direction of the reference layer at a position not overlapping with the first magnetization pinned layer in the stacking direction.

13. A magnetic array comprising a plurality of domain wall motion elements, each of the plurality of domain wall motion elements being the domain wall motion element according to claim 1.

Images