Magnetic multilayer film, magnetoresistive effect element, magnetic array, and neuromorphic device

Abstract

This magnetic multilayer film comprises a first ferromagnetic layer, a second ferromagnetic layer, and a nonmagnetic layer. The nonmagnetic layer is sandwiched between the first ferromagnetic layer and the second ferromagnetic layer in a layering direction. The nonmagnetic layer comprises a first region in contact with the first ferromagnetic layer, a second region in contact with the second ferromagnetic layer, and a third region sandwiched between the first region and the second region in the layering direction. The first region and the third region have different main elements of the respective constituent materials thereof. The second region and the third region have different main elements of the respective constituent materials thereof. At least one from among the first region, the second region, and the third region includes one of Ta, Mo, and W.

Description

Magnetic multilayer films, magnetoresistive elements, magnetic arrays, and neuromorphic devices

[0001] This disclosure relates to magnetic multilayer films, magnetoresistive elements, magnetic arrays, and neuromorphic devices.

[0002] A magnetoresistive element is a device in which the resistance value changes in the stacking direction due to the magnetoresistive effect. A magnetoresistive element comprises two ferromagnetic layers and a non-magnetic layer sandwiched between them. A magnetoresistive element in which a conductor is used for the non-magnetic layer is called a giant magnetoresistance (GMR) element, and a magnetoresistive element in which an insulating layer (tunnel barrier layer, barrier layer) is used for the non-magnetic layer is called a tunnel magnetoresistance (TMR) element. Magnetoresistive elements can be applied to a variety of uses, such as magnetic sensors, high-frequency components, magnetic heads, and non-volatile random access memory (MRAM).

[0003] For example, Patent Document 1 describes a magnetic domain wall-moving type magnetoresistive element (hereinafter referred to as a magnetic domain wall-moving element). A magnetic domain wall-moving element is an example of a magnetoresistive element. In a magnetic domain wall-moving element, the resistance value in the stacking direction changes depending on the position of the magnetic domain wall, and data can be recorded in multi-level or analog format. Magnetic domain wall-moving elements have high linearity and symmetry of resistance change, excellent rewrite endurance, and enable high-speed operation.

[0004] International Publication No. 2021 / 166892

[0005] The magnetoresistance curve of a magnetoresistive element exhibits a hysteresis loop. The greater the anisotropy of the magnetization of the ferromagnetic layer constituting the magnetoresistive element, the steeper the transition between the low-resistance state and the high-resistance state in the hysteresis loop. The steeper the transition between the low-resistance state and the high-resistance state of the magnetoresistive element, the higher the reliability of the data stored in the magnetoresistive element.

[0006] This disclosure has been made in view of the above-mentioned problems and aims to provide magnetic multilayer films, magnetoresistive elements, magnetic arrays, and neuromorphic devices with high magnetic anisotropy.

[0007] The magnetic multilayer film according to the first embodiment comprises a first ferromagnetic layer, a second ferromagnetic layer, and a non-magnetic layer. The non-magnetic layer is sandwiched between the first ferromagnetic layer and the second ferromagnetic layer in the stacking direction. The non-magnetic layer comprises a first region in contact with the first ferromagnetic layer, a second region in contact with the second ferromagnetic layer, and a third region sandwiched between the first region and the second region in the stacking direction. The first region and the third region have different main elements of their constituent materials. The second region and the third region have different main elements of their constituent materials. At least one of the first region, the second region, and the third region contains Ta, Mo, or W.

[0008] The magnetic multilayer film, magnetoresistive element, magnetic array, and neuromorphic device according to the above embodiment exhibit high magnetic anisotropy.

[0009] This is a block diagram of a magnetic array according to the first embodiment. This is a circuit diagram of the integrated region of the magnetic array according to the first embodiment. This is a cross-sectional view of the vicinity of the magnetic wall moving element of the magnetic array according to the first embodiment. This is a cross-sectional view of the magnetic wall moving element according to the first embodiment. This is a plan view of the magnetic wall moving element according to the first embodiment. This is a cross-sectional view of the magnetic wall moving layer according to the first embodiment. This is a cross-sectional view of the magnetic wall moving layer according to the first modified example. This is a schematic diagram of a neural network. This shows the magnetoresistance curves of the examples and comparative examples.

[0010] The following description of this embodiment will be made in detail with reference to the drawings as appropriate. The drawings used in the following description may be enlarged for convenience to clearly illustrate the features of this disclosure, and the dimensional ratios of each component may differ from those of the actual components. The materials, dimensions, etc., exemplified in the following description are examples only, and this disclosure is not limited to them. It is possible to modify and implement them as appropriate within the scope of achieving the effects of this disclosure.

[0011] First, directions are defined. The x-direction and the y-direction are directions substantially parallel to one surface of a substrate Sub (see FIG. 3) which will be described later. The x-direction is a direction in which a domain wall motion layer 10, which will be described later, extends. The y-direction is a direction orthogonal to the x-direction within the xy-plane. The z-direction is a direction from the substrate Sub, which will be described later, toward a domain wall motion element 100. In the present specification, the +z direction may be referred to as "upper" and the -z direction as "lower", but these expressions are for convenience and do not define the direction of gravity. In addition, in the present 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, the y-direction, and the z-direction. The same applies when extending in other directions. In addition, in the present specification, "connected" is not limited to the case of direct connection, but includes the case of connection via another object interposed therebetween.

[0012] [First Embodiment] FIG. 1 is a block diagram of a magnetic array MA according to the first embodiment. The magnetic array MA includes an integrated region 1 and a peripheral region 2. The magnetic array MA can be used for, for example, a magnetic memory, a product-sum arithmetic unit, a neuromorphic device, a spin memristor, and a magneto-optical element.

[0013] The integrated region 1 is a region in which a plurality of domain wall motion elements are integrated. The domain wall motion element is an example of a magnetoresistive effect element.

[0014] When the magnetic array MA is used as a memory, data is stored in the integrated region 1. When the magnetic array MA is used as a neuromorphic device, learning and inference are performed in the integrated region 1.

[0015] The peripheral region 2 is a region on which control elements that control the operation of the domain wall motion elements in the integrated region 1 are mounted. The peripheral region 2 includes, for example, a pulse applying device 3, a resistance detection device 4, and an output unit 5.

[0016] The pulse applying device 3 is configured to be capable of applying a pulse to at least one of the plurality of domain wall motion elements in the integrated region 1. The pulse applying device 3 includes, for example, a control unit 6 and a power supply 7.

[0017] The control unit 6 includes, for example, a processor and a memory. The processor is, for example, a CPU (Central Processing Unit). The processor operates based on a control program stored in the memory. The control unit 6 controls, for example, the address of a domain wall motion element to which a pulse is applied, the write pulse (voltage, pulse length) to be applied to a predetermined domain wall motion element, and the like. The control unit 6 may further include a clock, a counter, a random number generator, and the like. The clock serves as an indicator of the timing for applying a pulse, and the counter counts the number of times a pulse is applied, etc. The power supply 7 applies a pulse to the domain wall motion element in accordance with an instruction from the control unit 6.

[0018] The resistance detection device 4 is configured to be capable of detecting the resistance value of a domain wall motion element in the integrated region 1. The resistance detection device 4 may detect the resistance of each domain wall motion element in the integrated region 1, or may detect, for example, the total resistance of domain wall motion elements belonging to the same row / column. The resistance detection device 4 includes, for example, a comparator that compares the magnitude of detected resistance values. The comparator may, for example, compare detected resistance values with each other, or may compare a preset reference resistance value with the detected resistance value.

[0019] 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, and the like. When the magnetic array MA is used as a neuromorphic device, the output unit 5 may perform an operation of substituting the detection result of the resistance detection device 4 into an activation function. The operation is performed by, for example, the processor. The output unit 5 outputs the operation result to the outside. When the magnetic array MA is used as a neuromorphic device, for example, an operation such as outputting the operation result as an input signal to another magnetic array may be performed, or an operation such as outputting the operation result to the outside as an identification rate may be performed. The output unit 5 may also feed back the operation result to the pulse application device 3.

[0020] Figure 2 is a circuit diagram of the integrated region 1 according to the first embodiment. The integrated region 1 comprises a plurality of magnetic domain wall moving elements 100, a plurality of first wirings WL, a plurality of second wirings CL, a plurality of third wirings RL, a plurality of first switch elements SW1, and a plurality of second switch elements SW2. The third switch element SW3 may belong to, for example, the pulse application device 3 of the peripheral region 2.

[0021] The multiple magnetic domain wall moving elements 100 are arranged, for example, in a matrix. The multiple magnetic domain wall moving elements 100 are not limited to actual elements arranged in a matrix, but may also be arranged in a matrix in the circuit diagram.

[0022] Each of the first wirings WL is a write wiring. Each of the first wirings WL electrically connects the pulse application device 3 to one or more magnetic wall moving elements 100. Each of the second wirings CL is a common wiring that can be used both when writing and reading data. Each of the second wirings CL is connected, for example, to the resistance detection device 4. The second wirings CL may be provided on each of the multiple magnetic wall moving elements 100, or they may be provided across the multiple magnetic wall moving elements 100. Each of the third wirings RL is a read wiring. Each of the third wirings RL electrically connects the pulse application device 3 to one or more magnetic wall moving elements 100.

[0023] The first switch element SW1, the second switch element SW2, and the third switch element SW3 are elements that control the flow of current. The first switch element SW1, the second switch element SW2, and the third switch element SW3 are, for example, transistors. The first switch element SW1, the second switch element SW2, and the third switch element SW3 may be, for example, elements that utilize a phase change in the crystal layer, such as an ovonic threshold switch (OTS), elements that utilize a change in band structure, such as a metal-insulator transition (MIT) switch, elements that utilize a breakdown voltage, such as a Zener diode and an avalanche diode, or elements whose conductivity changes with a change in atomic position.

[0024] The first switch element SW1 and the second switch element SW2 are connected, for example, one to each magnetic domain wall moving element 100. The first switch element SW1 is connected, for example, between the magnetic domain wall moving element 100 and the first wiring WL. The second switch element SW2 is connected, for example, between the magnetic domain wall moving element 100 and the second wiring CL. The third switch element SW3 is connected, for example, across multiple magnetic domain wall moving elements 100. The third switch element SW3 is connected, for example, to the third wiring RL.

[0025] The positional relationship between the first switch element SW1, the second switch element SW2, and the third switch element SW3 is not limited to the case shown in Figure 2. For example, the first switch element SW1 may be connected across multiple magnetic domain wall moving elements 100 and located upstream of the first wiring WL. Also, for example, the second switch element SW2 may be connected across multiple magnetic domain wall moving elements 100 and located upstream of the second wiring CL. Also, for example, the third switch element SW3 may be connected to each of the magnetic domain wall moving elements 100.

[0026] Figure 3 is a cross-sectional view of the vicinity of a magnetic wall moving element 100 of the magnetic array MA according to the first embodiment. Figure 3 is a cross-section of one magnetic wall moving element 100 in Figure 2, cut by the xz plane passing through the center of the width in the y direction of the magnetic wall moving layer 10.

[0027] The first and second switching elements SW1 and SW2 shown in Figure 3 are transistors Tr. A transistor Tr has a gate electrode G, a gate insulating film GI, and a source S and drain D formed on a substrate Sub. The source S and drain D are determined by the direction of current flow and are in the same region. Figure 3 is just an example, and the positional relationship between the source S and drain D may be reversed. The substrate Sub is, for example, a semiconductor substrate. The third switching element SW3 is electrically connected to the third wiring RL and is, for example, located in a position shifted in the x-direction in Figure 3.

[0028] Each of the transistor Tr, the first wiring WL, the second wiring CL, the third wiring RL, and the magnetic wall moving element 100 is connected by via wiring V extending in the z direction or wiring W extending in either direction in the xy plane. The via wiring V and wiring W contain a conductive material. An insulating layer 90 is formed between different layers in the z direction, except for the via wiring V.

[0029] The insulating layer 90 is an insulating layer that insulates the spaces between wirings and elements in multilayer wiring. The magnetic domain wall moving 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.

[0030] Figure 4 is a cross-sectional view of the magnetic domain wall moving element 100, cut by the xz plane passing through the center of the magnetic domain wall moving layer 10 in the y direction. The arrows shown in the figure represent an example of the orientation direction of the magnetization of the ferromagnetic material. Figure 5 is a plan view of the magnetic domain wall moving element 100 as seen from the z direction.

[0031] The magnetic domain wall moving element 100 includes, for example, a magnetic domain wall moving layer 10, a spacer layer 20, a reference layer 30, a first magnetization fixed layer 40, and a second magnetization fixed layer 50. The magnetic domain wall moving layer 10 is an example of a free layer. The magnetic domain wall moving layer 10 is also called an analog layer or a magnetic recording layer.

[0032] The magnetic domain wall moving layer 10 extends in the x-direction. The magnetic domain wall moving layer 10 has multiple magnetic domains inside, and magnetic domain walls DW at the boundaries of the multiple magnetic domains. The magnetic domain wall moving layer 10 is a layer that can magnetically record information, for example, by a change in its magnetic state.

[0033] The domain wall moving layer 10 includes a magnetization fixed region A1, a magnetization fixed region A2, and a domain wall moving region A3. The magnetization fixed region A1 is a region overlapping the first magnetization fixed layer 40 when viewed from the z direction. The magnetization fixed region A2 is a region overlapping the second magnetization fixed layer 50 when viewed from the z direction. The domain wall moving region A3 is a region other than the magnetization fixed region A1 and the magnetization fixed region A2 of the domain wall moving layer 10. For example, the domain wall moving region A3 is a region sandwiched between the magnetization fixed region A1 and the magnetization fixed region A2 in the x direction.

[0034] The magnetization M of the magnetization fixed region A1 Ａ１ is fixed by the magnetization M of the first magnetization fixed layer 40 ４０ . The magnetization M of the magnetization fixed region A2 Ａ２ is fixed by the magnetization M of the second magnetization fixed layer 50 ５０ . The expression that magnetization is fixed means that the magnetization does not reverse in normal operation of the domain wall moving element 100 (when an unexpected external force is not applied). For example, the magnetization fixed region A1 and the magnetization fixed region A2 have opposite magnetization orientation directions.

[0035] In the domain wall moving region A3, the direction of magnetization can be changed, and this is a region where the domain wall DW can move. The domain wall moving region A3 includes a first magnetic domain A31 and a second magnetic domain A32. The first magnetic domain A31 and the second magnetic domain A32 have opposite magnetization orientation directions. A 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 oriented in the same direction as the magnetization M of the magnetization fixed region A1 Ａ１ , for example. The magnetization M of the second magnetic domain A32 Ａ３２ is oriented in the same direction as the magnetization M of the adjacent magnetization fixed region A2 Ａ２ , for example. In principle, the domain wall DW moves within the domain wall moving region A3 and does not enter the magnetization fixed region A1 or the magnetization fixed region A2.

[0036] When the ratio of the volumes of the first magnetic domain A31 and the second magnetic domain A32 within the magnetic domain wall movement region A3 changes, the magnetic domain wall DW moves. The magnetic domain wall DW moves by applying a writing current in the x direction of the magnetic domain wall movement region A3, or by applying an external magnetic field to the magnetic domain wall movement region A3. For example, when a writing current in the +x direction (e.g., a current pulse) is applied to the magnetic domain wall movement region A3, electrons flow in the opposite direction to the current, the -x direction, so the magnetic domain wall DW moves in the -x direction. When a current flows from the first magnetic domain A31 to the second magnetic domain A32, the electrons that have become spin-polarized in the second magnetic domain A32 move towards the magnetization M of the first magnetic domain A31. Ａ３１ The magnetization is reversed. Magnetization M of the first magnetic domain A31 Ａ３１ As it reverses, the magnetic domain wall DW moves in the -x direction.

[0037] Figure 6 is an enlarged view of the magnetic domain wall movement layer 10 according to the first embodiment. The magnetic domain wall movement layer 10 comprises, for example, a first ferromagnetic layer 11, a second ferromagnetic layer 12, and a non-magnetic layer 13. The first ferromagnetic layer 11 is in contact with, for example, a spacer layer 20. The second ferromagnetic layer 12 is in contact with, for example, a first magnetization fixed layer 40 and a second magnetization fixed layer 50. The first ferromagnetic layer 11 and the second ferromagnetic layer 12 are separated by a non-magnetic layer 13. The magnetic domain wall movement layer 10 is an example of a magnetic multilayer film.

[0038] The first ferromagnetic layer 11 and the second ferromagnetic layer 12 are magnetically coupled, and the magnetization direction of the first ferromagnetic layer 11 and the magnetization direction of the second ferromagnetic layer 12 change in conjunction. For example, if the magnetization direction of the magnetic domain wall movement region A3 of the second ferromagnetic layer 12 changes, the magnetization direction of the magnetic domain wall movement region A3 of the first ferromagnetic layer 11 also changes in accordance with that change in magnetization direction. Therefore, the magnetic domain walls DW of the magnetic domain wall movement layer 10 move in the same way in the first ferromagnetic layer 11 and the second ferromagnetic layer 12. The magnetic coupling between the first ferromagnetic layer 11 and the second ferromagnetic layer 12 may be either a ferromagnetic coupling or an antiferromagnetic coupling.

[0039] If the magnetic domain wall movement layer 10 has two ferromagnetic layers, a first ferromagnetic layer 11 and a second ferromagnetic layer 12, the roles of each ferromagnetic layer can be divided. For example, the magnetic domain wall movement element 100 is required to have a high MR ratio and to be able to precisely control the magnetic domain wall DW. The first ferromagnetic layer 11 faces the reference layer 30 with a spacer layer 20 in between. The first ferromagnetic layer 11 has a greater influence on the MR ratio of the magnetic domain wall movement element 100 than the second ferromagnetic layer 12. Therefore, a material that can achieve a high MR ratio is applied to the first ferromagnetic layer 11. On the other hand, a material that makes it easy to precisely control the movement of the magnetic domain wall DW can also be applied to the second ferromagnetic layer 12. In this way, by composing the magnetic domain wall movement layer 10 with multiple ferromagnetic layers and dividing the role of each ferromagnetic layer, the overall characteristics of the magnetic domain wall movement layer 10 can be improved.

[0040] The first ferromagnetic layer 11 is composed of a magnetic material. The first ferromagnetic layer 11 is, 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, an alloy containing these metals and at least one of the elements B, C, and N, etc. The first ferromagnetic layer 11 is, for example, Co ｘ Fe １－ｘ B ｙ Compounds expressed as (0 ≤ x ≤ 1, 0 < y ≤ 35) are also acceptable. These compounds tend to have a bcc crystal structure, which facilitates lattice matching with the spacer layer 20 and can increase the MR ratio of the magnetic domain wall moving element 100.

[0041] The second ferromagnetic layer 12 is composed of a magnetic material. The second ferromagnetic layer 12 may have at least one element selected from the group consisting of Co, Ni, Fe, Pt, Pd, Gd, Tb, Mn, Ge, and Ga. The second ferromagnetic layer 12 may be, for example, a Co-Ni laminate, a Co-Pt laminate, a Co-Pd laminate, a MnGa-based material, a GdCo-based material, or a TbCo-based material. Ferrimagnetic materials such as MnGa-based materials, GdCo-based materials, and TbCo-based materials have low saturation magnetization, which reduces the threshold current required to move the magnetic domain wall DW. The second ferromagnetic layer 12 may also be a Co-Ni laminate, a Co-Pt laminate, or a Co-Pd laminate. These have high coercivity, which slows down the movement speed of the magnetic domain wall DW. The second ferromagnetic layer 12 may have a crystal structure such as an fcc structure or an hcp structure. The crystal structure of the second ferromagnetic layer 12 may be different from the crystal structure of the first ferromagnetic layer 11.

[0042] The non-magnetic layer 13 has a first region 131, a second region 132, and a third region 133 in the z direction. The first region 131 is in contact with the first ferromagnetic layer 11. The second region 132 is in contact with the second ferromagnetic layer 12. The third region 133 is sandwiched between the first region 131 and the second region 132 in the z direction.

[0043] The first region 131, the second region 132, and the third region 133 are each composed of non-magnetic elements. The first region 131 and the third region 133 have different main elements in their constituent materials. The second region 132 and the third region 133 have different main elements in their constituent materials. The first region 131 and the second region 132 may have the same main element. The main element is the element that accounts for 50% or more of the material constituting that region. The elements constituting the non-magnetic layer 13 can be analyzed by performing elemental mapping using energy-dispersive X-ray spectroscopy with a transmission electron microscope or scanning electron microscope.

[0044] The concentration of the main element constituting the first region 131 may change continuously within the non-magnetic layer 13. For example, upon reaching the third region 133, the main element constituting the non-magnetic layer 13 switches from the main element constituting the first region 131 to the main element constituting the third region. At this time, the main element constituting the first region 131 does not need to be completely zero; it may be included in the second region 132 and the third region 133. For example, the concentration of the main element constituting the first region 131 may be maximum in the first region 131, and then decrease in the order of the third region 133 and the second region 132.

[0045] Similarly, the concentration of the main element constituting the second region 132 may change continuously within the non-magnetic layer 13. For example, the concentration of the main element constituting the second region 132 may be maximum in the second region 132, and then decrease in the third region 133 and the first region 131.

[0046] Similarly, the concentration of the main element constituting the third region 133 may change continuously within the non-magnetic layer 13. For example, the concentration of the main element constituting the third region 133 may be maximum in the third region 133 and lower in the first region 131 and the second region 132 than in the third region 133. For example, when the main element constituting the third region 133 is the first element, the concentration of the first element on the first surface S1 in contact with the first ferromagnetic layer 11 of the first region 131 may be lower than the concentration of the first element on the second surface S2 of the first region 131. Also, the concentration of the first element may gradually decrease from the second surface S2 toward the first surface S1.

[0047] At least one of the first region 131, the second region 132, and the third region 133 contains Ta, Mo, or W. At least one of the first region 131, the second region 132, and the third region 133 may contain Ta, Mo, or W as the main element. Ta, Mo, and W maintain the magnetic coupling between the first ferromagnetic layer 11 and the second ferromagnetic layer 12 without breaking it. Also, Ta, Mo, and W readily adopt a bcc-type crystal structure and readily break the crystalline continuity between the first ferromagnetic layer 11 and the second ferromagnetic layer 12. When the crystallinity between them is broken, the crystal structure of the second ferromagnetic layer 12 becomes less affected by the crystal structure of the first ferromagnetic layer 11, and strain is less likely to occur within the second ferromagnetic layer 12. When the crystallinity of the second ferromagnetic layer 12 increases, the magnetic anisotropy of the domain wall migration layer 10 increases.

[0048] The main elements constituting the third region 133 are, for example, Ta, Mo, or W. When these main elements are positioned in the middle of the z-direction, the perpendicular magnetic anisotropy of the first ferromagnetic layer 11 and the second ferromagnetic layer 12 is particularly high. This is because the first ferromagnetic layer 11 and the second ferromagnetic layer 12 become less susceptible to each other's crystal structure influences.

[0049] The main elements constituting the first region 131 and the second region 132 are, for example, Ni, Cu, Zn, Ru, Pd, Ag, Ir, Pt, and Au. The main elements constituting the first region 131 and the second region 132 may also be, for example, Ru, Pd, and Pt. These elements make it easy to select either an fcc structure or an hcp structure.

[0050] When the first ferromagnetic layer 11 has a bcc structure and the first region 131 is composed of an element that readily selects an fcc or hcp structure, the first region 131 reduces the influence of the crystal structure of the first ferromagnetic layer 11 on the crystallinity of the layer deposited above the first region 131. By distorting the crystal in the first region 131 and then breaking the continuity of the crystal in the third region 133, the crystal structure of the second ferromagnetic layer 12 becomes less susceptible to the influence of the crystal structure of the first ferromagnetic layer 11.

[0051] Furthermore, when the second ferromagnetic layer 12 has an FCC structure or an HCP structure, and the second region 132 has an FCC structure or an HCP structure, the crystallinity of the second ferromagnetic layer 12 tends to increase. This is because the second region 132 functions as a seed for crystal growth in the second ferromagnetic layer 12.

[0052] In the non-magnetic layer 13, the thicknesses of the first region 131, the second region 132, and the third region 133 are not particularly limited. For example, the thickness of the second region 132 may be greater than that of the first region 131. By increasing the thickness of the second region 132, the laminated surface on which the second ferromagnetic layer 12 is laminated can be made flat. When the interface between the second ferromagnetic layer 12 and the non-magnetic layer 13 becomes flat, the magnetic anisotropy of the second ferromagnetic layer 12 increases due to interfacial magnetic anisotropy. The thicknesses of the first region 131, the second region 132, and the third region 133 can be determined by elemental mapping using EDX, measuring the thickness until the dominant element switches.

[0053] The thickness of the non-magnetic layer 13 may be, for example, 6 Å or more and 16 Å or less. If the thickness of the non-magnetic layer 13 is thin, the magnetic coupling between the first ferromagnetic layer 11 and the second ferromagnetic layer 12 becomes stronger. Alternatively, the thickness of the non-magnetic layer 13 may be, for example, greater than 16 Å and 30 Å or less. By creating a three-layer structure with different main elements in the stacking direction as described above, the magnetic coupling between the first ferromagnetic layer 11 and the second ferromagnetic layer 12 can be maintained even if the thickness of the non-magnetic layer 13 is thick. Furthermore, if the thickness of the non-magnetic layer 13 is greater than 16 Å and 30 Å or less, the first ferromagnetic layer 11 and the second ferromagnetic layer 12 become antiferromagnetically coupled.

[0054] Figure 7 is an enlarged view of the magnetic domain wall moving layer 10A according to the first modified example. The magnetic domain wall moving layer 10A differs from the magnetic domain wall moving layer 10 in that it further has an intermediate layer 14. The magnetic domain wall moving layer 10A can be used in place of the magnetic domain wall moving layer 10. The magnetic domain wall moving layer 10A is an example of a magnetic laminated film. In the magnetic domain wall moving layer 10A, components similar to those in the magnetic domain wall moving layer 10 are denoted by the same reference numerals and their explanation is omitted.

[0055] The intermediate layer 14 is, for example, sandwiched between the first ferromagnetic layer 11 and the non-magnetic layer 13 in the z-direction. The intermediate layer 14 contains, for example, one or more materials selected from the group consisting of Ti, V, Cr, Mo, Ra, and W.

[0056] The elements constituting the intermediate layer 14 adsorb boron when heated. The concentration of boron in the intermediate layer 14 may be higher than, for example, the concentration of boron in the first ferromagnetic layer 11. For example, if the first ferromagnetic layer 11 contains boron, the intermediate layer 14 prevents the boron from diffusing to the non-magnetic layer 13 and the second ferromagnetic layer 12. Boron diffused from the first ferromagnetic layer 11 can disrupt the crystal structure of the second ferromagnetic layer 12 and cause a decrease in the magnetic anisotropy of the second ferromagnetic layer 12.

[0057] The spacer layer 20 (see Figure 4) is located between the magnetic domain wall moving layer 10 and the reference layer 30. The spacer layer 20 is laminated on one surface of the reference layer 30.

[0058] The spacer layer 20 is made of, for example, a non-magnetic insulator, semiconductor, or metal. A non-magnetic insulator is, for example, Al ２ O ３ SiO ２ MgO, MgAl ２ O ４ These are materials in which some of the Al, Si, and Mg are replaced with Zn, Be, etc. These materials have a large band gap and excellent insulating properties. When the spacer layer 20 is made of a non-magnetic insulator, the spacer layer 20 is a tunnel barrier layer. Non-magnetic metals include, for example, Cu, Au, Ag, etc. Non-magnetic semiconductors include, for example, Si, Ge, CuInSe ２ CuGaSe ２ , Cu(In,Ga)Se ２ And so on.

[0059] The thickness of the spacer layer 20 is, for example, 20 Å or more, and may also be 25 Å or more. A thicker spacer layer 20 increases the resistive area area (RA) of the magnetic domain wall moving element 100. The resistive area area area (RA) of the magnetic domain wall moving element 100 is 1 × 10⁻¹⁶. ４ Ωμm ２ Preferably, it is 5 x 10 ４ Ωμm ２ The above is more preferable. The resistive area product (RA) of the magnetic wall moving element 100 is expressed as the product of the element resistance of one magnetic wall moving element 100 and the element cross-sectional area of ​​the magnetic wall moving element 100 (the area of ​​the cross-section obtained by cutting the spacer layer 20 in the xy plane).

[0060] The reference layer 30, together with the magnetic domain wall movement layer 10, is sandwiched between the spacer layer 20. At least a portion of the reference layer 30 is positioned to overlap with the magnetic domain wall movement layer 10 in the z-direction. The magnetization of the reference layer 30 is less likely to reverse than the magnetization of the magnetic domain wall movement region A3 of the magnetic domain wall movement layer 10. The magnetization of the reference layer 30 remains fixed and does not change direction when an external force sufficient to reverse the magnetization of the magnetic domain wall movement region A3 is applied. The reference layer 30 is sometimes referred to as a fixed layer.

[0061] The reference layer 30 includes, for example, a ferromagnetic material. The reference layer is Mn ３ An antiferromagnetic material such as Sn may also be used. The reference layer 30 includes, for example, a material that readily exhibits a coherent tunneling effect with respect to the first ferromagnetic layer 11. The reference layer 30 includes, 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, an alloy containing these metals and at least one of the elements B, C, and N, etc. The reference layer 30 is, for example, Co-Fe, Co-Fe-B, or Ni-Fe.

[0062] The reference layer 30 may be, for example, a Heusler alloy. Heusler alloys are half-metallic and have a high spin polarizability. Heusler alloys are XYZ or X ２ It is an intermetallic compound with the chemical composition YZ, where X is a transition metal or noble metal element of group Co, Fe, Ni, or Cu on the periodic table, Y is a transition metal or element species of group Mn, V, Cr, or Ti, and Z is a typical element of group III to V. For example, a Heusler alloy is Co 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 These are some examples.

[0063] The first magnetization fixed layer 40 and the second magnetization fixed layer 50 are connected to the magnetic domain wall moving layer 10. The first magnetization fixed layer 40 and the second magnetization fixed layer 50 are connected to different positions on the magnetic domain wall moving layer 10. The first magnetization fixed layer 40 and the second magnetization fixed layer 50 are spaced apart in the x-direction. The first magnetization fixed layer 40 fixes the magnetization of magnetization fixed region A1. The second magnetization fixed layer 50 fixes the magnetization of magnetization fixed region A2.

[0064] The first magnetization fixed layer 40 and the second magnetization fixed layer 50 are, for example, ferromagnetic materials. The first magnetization fixed layer 40 and the second magnetization fixed layer 50 can be made of the same materials as the first ferromagnetic layer 11, the second ferromagnetic layer 12, and the reference layer 30.

[0065] The first magnetization fixed layer 40 and the second magnetization fixed layer 50 are not limited to ferromagnetic materials. If the first magnetization fixed layer 40 and the second magnetization fixed layer 50 are not ferromagnetic materials, the current density of the current flowing through the magnetic domain wall moving layer 10 changes rapidly in the region overlapping with the first magnetization fixed layer 40 or the second magnetization fixed layer 50, thereby restricting the movement of the magnetic domain wall DW and fixing the magnetization of the magnetization fixed region A1 and the magnetization fixed region A2.

[0066] The magnetic domain wall moving element 100 may have layers other than the magnetic domain wall moving layer 10, the spacer layer 20, and the reference layer 30. For example, a magnetic layer may be provided on the side of the reference layer 30 opposite to the spacer layer 20, via a magnetic coupling layer. The reference layer 30, the magnetic coupling layer, and the magnetic layer form a synthetic antiferromagnetic structure (SAF structure). The synthetic antiferromagnetic structure consists of two magnetic layers sandwiching a non-magnetic layer. The coercivity of the reference layer 30 is increased by the antiferromagnetic coupling of the reference layer 30 compared to the case without a magnetic layer. The magnetic layer may include, for example, a ferromagnetic material and an antiferromagnetic material such as IrMn or PtMn. The magnetic coupling layer may include, for example, at least one selected from the group consisting of Ru, Ir, and Rh.

[0067] The direction of magnetization in each layer of the magnetic domain wall moving element 100 can be confirmed, for example, by measuring the magnetization curve. The magnetization curve can be measured, for example, using MOKE (Magneto-Optical Kerr Effect). MOKE measurement is a measurement method that uses the magneto-optical effect (magnetic Kerr effect), which occurs when linearly polarized light is incident on the object to be measured and the direction of polarization is rotated.

[0068] The magnetic domain wall moving element 100 is formed by a layer stacking process and a processing process in which a portion of each layer is processed into a predetermined shape. Layer stacking can be performed using sputtering, chemical vapor deposition (CVD), electron beam deposition (EB deposition), atomic laser deposition, etc. Processing of each layer can be performed using photolithography and etching (e.g., Ar etching), etc.

[0069] The non-magnetic layer 13 is divided into three regions in the z direction by changing the type of element sputtered in each region.

[0070] Next, we will explain the operation of writing signals to the magnetic array MA and the operation of reading signals from the magnetic array MA.

[0071] The operation of writing signals to the magnetic array MA will now be described. The writing operation is performed by applying a write pulse to the magnetic domain wall moving element 100. This process of applying the write pulse is performed, for example, by the processor executing an application program stored in the control unit 6.

[0072] First, select the domain wall moving element 100 to which the pulse is applied. When the magnetic array MA is used as a magnetic memory, the domain wall moving element 100 to which the pulse is applied is the element that stores data. When the magnetic array MA is used as a neural network, the domain wall moving element 100 to which the pulse is applied is the element that performs learning.

[0073] The control unit 6 controls which of the multiple magnetic domain wall moving elements 100 to which a pulse is applied, according to the control program. The control unit 6 turns on the first switch element SW1 and the second switch element SW2 connected to the magnetic domain wall moving element 100 to which the pulse is applied, and turns off the first switch element SW1 and the second switch element SW2 connected to the other magnetic domain wall moving elements 100.

[0074] Then, a write pulse is applied to the magnetic domain wall moving element 100 to be written to. The write pulse may be a square wave, a spike wave, or any other waveform. When the write pulse is applied to the magnetic domain wall moving element 100, the position of the magnetic domain wall DW changes, and the conductance or resistance of the magnetic domain wall moving element 100 changes. This conductance or resistance of the magnetic domain wall moving element 100 becomes the written data.

[0075] As described above, the magnetic array MA completes its writing operation when a computer executes the operating program.

[0076] Next, the operation of reading signals from the magnetic array MA will be described. The reading operation is performed, for example, by the processor executing a control program stored in the control unit 6.

[0077] First, the pulse application device 3 selects a domain wall moving element 100 to which a read pulse is applied, according to the control program. When the magnetic array MA is used as a magnetic memory, the domain wall moving element 100 to which the read pulse is applied is the element that reads the data. When the magnetic array MA is used as a neural network, the application of a read pulse to a predetermined domain wall moving element 100 corresponds to the product 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 the discrimination operation of the neural network.

[0078] The control unit 6 controls which of the multiple magnetic wall moving elements 100 to which a pulse is applied. The control unit 6 turns on the third switch element SW3 and the second switch element SW2 connected to the magnetic wall moving element 100 to which the pulse is applied, and turns off the third switch element SW3 and the second switch element SW2 connected to the other magnetic wall moving elements 100.

[0079] Next, the pulse application device 3 applies a readout pulse to a predetermined magnetic domain wall moving element 100 according to the control program. The readout pulse is applied, for example, between the reference layer 30 and the second magnetization fixed layer 50. The voltage of the readout pulse is such that a current density less than the critical current density required to move the magnetic domain wall DW of the magnetic domain wall moving layer 10 is obtained. In other words, the readout pulse does not move the magnetic domain wall DW.

[0080] The resistance detection device 4 detects the resistance value of the magnetic wall moving element 100 to which a readout pulse has been applied. The output unit 5 outputs the calculation result to the outside, for example.

[0081] The magnetic domain wall moving element 100 according to this embodiment has excellent magnetic properties because its free layer is made of a predetermined magnetic multilayer film. Specifically, the magnetoresistance curve of the magnetic domain wall moving element 100 shows a steep transition between the low-resistance state and the high-resistance state in the hysteresis loop. The magnetic domain wall moving element 100, which shows a steep transition between the low-resistance state and the high-resistance state of the magnetoresistance effect element, has high data reliability. The reason why the transition between the low-resistance state and the high-resistance state of the magnetoresistance effect element is steep is because the magnetic anisotropy of the free layer is large.

[0082] The magnetic array MA according to this embodiment can be applied to neuromorphic devices.

[0083] A neuromorphic device, for example, comprises a magnetic array MA and an output conversion unit. The output conversion unit has an activation function. The output conversion unit consists of a resistance detection device 4 and an output unit 5. The output conversion unit converts the sum-of-products calculation result output from the second wiring CL according to the activation function.

[0084] Neuromorphic devices are devices that perform neural network calculations. They artificially mimic the relationship between neurons and synapses in the human brain.

[0085] Figure 8 is a schematic diagram of a neural network (NN). A neural network (NN) has an input layer L ｉｎ and the middle layer L ｍ and output layer L ｏｕｔ It has the following. In Figure 8, the intermediate layer L ｍ The example shows three layers, but the intermediate layer L ｍ The number is irrelevant. Input layer L ｉｎ and the middle layer L ｍ and output layer L ｏｕｔ Each of these has multiple chips C, and each chip C corresponds to a neuron in the brain. Input layer L ｉｎ and the middle layer L ｍ and output layer L ｏｕｔ Each of these is connected by a transmission mechanism. The transmission mechanism corresponds to a synapse in the brain. The neural network (NN) improves its accuracy in answering problems by having the transmission mechanism (synapse) learn. Learning is the process of finding knowledge from information that may be useful in the future. The neural network (NN) learns by operating while changing the weights applied to the transmission mechanism. The transmission mechanism performs a multiplication operation by multiplying the input signal by a weight, and a sum operation by adding the results of the multiplication operation. In other words, the transmission mechanism performs a sum-of-products operation.

[0086] The magnetic array MA can perform sum-of-products operations. The resistance of the magnetic domain wall moving element 100 changes in a multi-level or analog manner as the position of the magnetic domain wall DW changes. Designing the resistance of the magnetic domain wall moving element 100 and its reciprocal, the conductance, corresponds to assigning weights to the transmission means.

[0087] For example, in Figure 2, current is passed from the third wiring RL to the second wiring CL. The current (output value) output from the second wiring CL varies depending on the conductance (weight) of the magnetic domain wall moving element 100. In other words, applying current from the third wiring RL to the second wiring CL corresponds to the multiplication operation in a neural network NN. Furthermore, the second wiring CL is connected to multiple magnetic domain wall moving elements 100 belonging to the same column, and the current detected at the end of the second wiring CL is the sum of the results of the multiplication operation performed by each magnetic domain wall moving element 100. Therefore, the magnetic array MA functions as a multiply-accumulate unit for neuromorphic devices.

[0088] Although preferred embodiments of this disclosure have been described in detail above, this disclosure is not limited to these embodiments. For example, characteristic configurations of each embodiment may be combined, or parts may be modified without altering the essence of the invention.

[0089] For example, while we have shown an example where the magnetoresistive element is a domain wall moving element, the configuration of the magnetoresistive element is not limited to this example. For instance, the magnetic multilayer film may be applied to the free layer of a spin-transfer torque (STT) type magnetoresistive element, or to the free layer of a spin-orbit torque (SOT) type magnetoresistive element.

[0090] (Example 1) In Example 1, a magnetoresistive element was fabricated in which a reference layer, a spacer layer, a first ferromagnetic layer, a non-magnetic layer, and a second ferromagnetic layer were stacked in order from the side closest to the substrate. The first ferromagnetic layer and the second ferromagnetic layer are magnetically coupled, and the first ferromagnetic layer, the non-magnetic layer, and the second ferromagnetic layer together function as a free layer.

[0091] The composition of each layer is shown below. Reference layer: A multilayer film in which 0.4 nm Co and 0.4 nm Pt are alternately stacked, with 0.2 nm W and 1.1 nm CoFeB stacked in sequence. Spacer layer: Mg-Al-O, 2.5 nm First ferromagnetic layer: (Co ０．２ Fe ０．８ ) ０．２ B ０．８1.4 nm non-magnetic layer: A three-layer structure consisting of a first region, a third region, and a second region, starting from the side closest to the first ferromagnetic layer. First region: Ru, 2 nm Third region: Ta, 2 nm Second region: Pt, 8 nm Second ferromagnetic layer: A multilayer film in which 0.4 nm Co and 0.4 nm Pt are alternately stacked.

[0092] Next, the magnetoresistance curve of the magnetoresistive element of Example 1 was measured. The magnetoresistance curve was determined by measuring the change in the resistance value of the magnetoresistive element by monitoring the applied voltage to the magnetoresistive element with a voltmeter while sweeping a magnetic field across the magnetoresistive element from an external source, with a constant current flowing in the stacking direction of the magnetoresistive element. Figure 9 shows the magnetoresistance curve of Example 1. The horizontal axis represents the magnetic field applied to the magnetoresistive element, and the vertical axis represents the resistance value in the stacking direction of the magnetoresistive element.

[0093] (Comparative Example 1) The magnetoresistive element according to Comparative Example 1 differs from that of Example 1 in that the non-magnetic layer consists of two layers of Ta and Pt, in that order from the side closest to the substrate. The magnetoresistive curve of the magnetoresistive element of Comparative Example 1 was measured in the same manner as in Example 1. Figure 9 shows the magnetoresistive curve of Comparative Example 1.

[0094] Comparing Example 1 with Comparative Example 1, it can be seen that the magnetoresistance curve shown in Figure 9 exhibits a steeper transition between the low-resistance state and the high-resistance state in the hysteresis loop.

[0095] (Examples 2 and 3) Example 2 differs from Example 1 in that the element constituting the third region is changed. In Example 2, the element constituting the third region is Mo, and in Example 3, the element constituting the third region is W. Similar magnetoresistance curves were measured in Examples 2 and 3. The magnetoresistance curves of Examples 2 and 3 were in close agreement with the magnetoresistance curve of Example 1. The magnetoresistance curves of Examples 2 and 3 showed a steeper transition between the low-resistance state and the high-resistance state than Comparative Example 1.

[0096] (Examples 4-11) Examples 4-11 differ from Example 1 in that the elements constituting the first region are changed. The elements constituting the first region in each example are shown below. Example 4: Ni Example 5: Cu Example 6: Zn Example 7: Pd Example 8: Ag Example 9: Ir Example 10: Pt Example 11: Au The magnetoresistance curve was measured in each example. The magnetoresistance curves of Examples 4-11 were substantially the same. Figure 9 shows the magnetoresistance curve of Example 4 as a representative example. The magnetoresistance curves of Examples 4-11 showed a steeper transition between the low-resistance state and the high-resistance state compared to Comparative Example 1. Also, when the first region was Ru (Example 1), Pd (Example 7), and Pt (Example 10), there was a tendency for the transition between the low-resistance state and the high-resistance state to be steeper compared to the other examples.

[0097] (Examples 12-19) Examples 12-19 differ from Example 1 in that the elements constituting the third region are changed. The elements constituting the first region in each example are shown below. Example 12: Ni Example 13: Cu Example 14: Zn Example 15: Pd Example 16: Ag Example 17: Ir Example 18: Ru Example 19: Au The magnetoresistance curve was measured in each example. The magnetoresistance curves of Examples 12-19 were substantially the same. Figure 9 shows the magnetoresistance curve of Example 12 as a representative example. The magnetoresistance curves of Examples 12-19 showed a steeper transition between the low-resistance state and the high-resistance state compared to Comparative Example 1. Also, when the third region was Pt (Example 1), Pd (Example 7), and Ru (Example 10), there was a tendency for the transition between the low-resistance state and the high-resistance state to be steeper compared to the other examples.

[0098] (Examples 20-25) Examples 20-25 differ from Example 1 in that an intermediate layer is provided between the first ferromagnetic layer and the non-magnetic layer. The elements constituting the intermediate layer in each example are shown below. Example 20: Ti Example 21: V Example 22: Cr Example 23: Mo Example 24: Ra Example 25: W The magnetoresistance curve was measured in each example. The magnetoresistance curves of Examples 20-25 were substantially the same. Figure 9 shows the magnetoresistance curve of Example 20 as a representative example. The magnetoresistance curves of Examples 20-25 showed a steeper transition between the low-resistance state and the high-resistance state than Comparative Example 1. Furthermore, the provision of an intermediate layer tended to make the transition between the low-resistance state and the high-resistance state steeper than in Example 1.

[0099] (Examples 26-28) Examples 26-28 differ from Example 1 in that the material constituting the third region is Pt, and the element constituting the second region is changed. The elements constituting the second region in each example are shown below. Example 26: Ta Example 27: Mo Example 28: W The magnetoresistance curve was measured in each example. The magnetoresistance curves of Examples 26-28 were substantially the same. Figure 9 shows the magnetoresistance curve of Example 26 as a representative example. The magnetoresistance curves of Examples 26-28 showed a steeper transition between the low-resistance state and the high-resistance state than Comparative Example 1.

[0100] (Examples 29-31) Examples 29-31 differ from Example 1 in that the material constituting the third region is Ru, and the element constituting the first region is changed. The elements constituting the first region in each example are shown below. Example 29: Ta Example 30: Mo Example 31: W The magnetoresistance curve was measured in each example. The magnetoresistance curves of Examples 29-31 were substantially the same. Figure 9 shows the magnetoresistance curve of Example 29 as a representative example. The magnetoresistance curves of Examples 29-31 showed a steeper transition between the low-resistance state and the high-resistance state than Comparative Example 1.

[0101] (Examples 32-34) Example 32 differs from Example 1 in that the elements of each region constituting the non-magnetic layer were changed as follows: Example 32: Region 1 = Mo, Region 3 = Ta, Region 2 = W Example 33: Region 1 = W, Region 3 = Ta, Region 2 = W Example 34: Region 1 = Mo, Region 3 = Ta, Region 2 = Mo Magnetoresistance curves were measured in Examples 32-34. The magnetoresistance curves of Examples 32-34 were substantially the same. Figure 9 shows the magnetoresistance curve of Example 32 as a representative example. The magnetoresistance curves of Examples 32-34 showed a steeper transition between the low-resistance state and the high-resistance state than Comparative Example 1.

[0102] (Example 35) Example 35 differs from Example 1 in that the elements of each region constituting the non-magnetic layer were changed as follows, and the thickness of each region was changed. Region 1 = W, 5 nm Region 3 = Ta, 5 nm Region 2 = W, 10 nm In Example 35, the magnetoresistance curve was measured. Figure 9 shows the magnetoresistance curve of Example 35. The magnetoresistance curve of Example 35 showed a steeper transition between the low-resistance state and the high-resistance state than that of Comparative Example 1.

[0103] 1. Integrated region 2. Peripheral region 3. Pulse application device 4. Resistance detection device 5. Output unit 6. Control unit 7. Power supply 10, 10A 11. Magnetic wall moving layer 12. First ferromagnetic layer 13. Second ferromagnetic layer 14. Non-magnetic layer 14. Intermediate layer 20. Spacer layer 30. Reference layer 40. First magnetization fixed layer 50. Second magnetization fixed layer 90. Insulating layer 100. Magnetic wall moving element 131. First region 132. Second region 133. Third region A1, A2. Magnetization fixed region A3. Magnetic wall moving region A31. First magnetic domain A32. Second magnetic domain DW. Magnetic wall MA. Magnetic array S1. First surface S2. Second surface

Claims

1. A magnetic laminate film comprising a first ferromagnetic layer, a second ferromagnetic layer, and a non-magnetic layer, wherein the non-magnetic layer is sandwiched between the first ferromagnetic layer and the second ferromagnetic layer in the stacking direction, the non-magnetic layer comprises a first region in contact with the first ferromagnetic layer, a second region in contact with the second ferromagnetic layer, and a third region sandwiched between the first region and the second region in the stacking direction, wherein the first region and the third region have different main elements of the constituent material, the second region and the third region have different main elements of the constituent material, and at least one of the first region, the second region and the third region contains Ta, Mo, or W.

2. The magnetic multilayer film according to claim 1, wherein the third region includes any of Ta, Mo, or W.

3. The magnetic laminate according to claim 1, wherein the thickness of the second region is greater than the thickness of the first region.

4. The layer further comprises an intermediate layer sandwiched between the first ferromagnetic layer and the non-magnetic layer in the stacking direction, wherein the first ferromagnetic layer is made of Co ｘ Fe １－ｘ B ｙ The magnetic laminated film according to claim 1, expressed as (0 ≤ x ≤ 1, 0 < y ≤ 35), wherein the intermediate layer contains one or more selected from the group consisting of Ti, V, Cr, Mo, Ra, and W.

5. The magnetic laminated film according to claim 4, wherein the intermediate layer contains boron, and the concentration of boron in the intermediate layer is higher than the concentration of boron in the first ferromagnetic layer.

6. The magnetic multilayer film according to claim 1, wherein the first region contains the first element constituting the third region, and the concentration of the first element on the first surface of the first region in contact with the first ferromagnetic layer is lower than the concentration of the first element on the second surface of the first region facing the first surface.

7. The magnetic multilayer film according to claim 1, wherein the second ferromagnetic layer has a crystalline structure of fcc or hcp.

8. The magnetic multilayer film according to claim 1, wherein the concentration of the principal element constituting each region changes continuously between the first region and the third region, and between the second region and the third region.

9. The magnetic laminated film according to claim 1, wherein the thickness of the non-magnetic layer is 6 Å or more and 16 Å or less.

10. The magnetic laminated film according to claim 1, wherein the thickness of the non-magnetic layer is greater than 16 Å and less than or equal to 30 Å.

11. A magnetoresistive element comprising a free layer, a reference layer, and a spacer layer sandwiched between the free layer and the reference layer, wherein the free layer is the magnetic multilayer film described in claim 1.

12. The magnetoresistive element according to claim 11, wherein the free layer has a magnetic domain wall inside.

13. A magnetic array comprising the magnetoresistive element described in claim 11.

14. A neuromorphic device comprising the magnetoresistive element described in claim 11.

Images