CN111183480A

CN111183480A - Composite free layer for magnetoresistive random access memory

Info

Publication number: CN111183480A
Application number: CN201880061542.6A
Authority: CN
Inventors: Y-S·崔
Original assignee: SanDisk Technologies LLC
Current assignee: SanDisk Technologies LLC
Priority date: 2017-11-21
Filing date: 2018-09-28
Publication date: 2020-05-19
Anticipated expiration: 2038-09-28
Also published as: CN111183480B; DE112018004229T5; WO2019103788A1

Abstract

The invention discloses a device, a system and a method for a magnetoresistive random access memory. A magnetic tunnel junction (350) for storing data includes a fixed layer (412), a barrier layer (410), and a composite free layer (400). The barrier layer (410) is an OS disposed between the fixed layer (412) and the composite free layer (400). The composite free layer (400) includes a ferromagnetic amorphous layer (406) and an in-plane anisotropic free layer (408). A Spin Hall Effect (SHE) layer (402) may be coupled to the composite free layer (400) of the magnetic tunnel junction (350). The SHE layer (402) may be configured such that in-plane currents within the SHE layer (402) generate spin currents in the composite free layer (400).

Description

Composite free layer for magnetoresistive random access memory

Technical Field

In various embodiments, the present disclosure relates to magnetoresistive random access memories, and more particularly to a composite free layer for a magnetoresistive random access memory.

Background

Various types of Magnetoresistive Random Access Memories (MRAMs) use magnetic tunnel junctions to store data. A Magnetic Tunnel Junction (MTJ) may include "fixed" and "free" magnetic layers, where the magnetic moment of the free layer may be switched to be parallel or anti-parallel to the magnetic moment of the fixed layer. A thin dielectric or barrier layer may separate the fixed and free layers, and current may flow through the barrier layer due to quantum tunneling. The difference in resistance between the parallel and anti-parallel states allows data to be stored. For example, a low resistance may correspond to a binary "1" and a high resistance may correspond to a binary "0". Alternatively, a low resistance may correspond to a binary "0" and a high resistance may correspond to a binary "1".

In Spin Transfer Torque (STT) MRAM, data may be written by passing a spin-polarized current through the MTJ to change the magnetic moment of the free layer. However, high write currents through the MTJ may accelerate wear of the barrier layer, and spin-polarized read currents may disturb or change the stored data. In contrast, in Spin Orbit Torque (SOT) MRAM, data can be written by applying a current through the spin hall effect material adjacent to the free layer, thus generating a pure spin current that changes the magnetic moment of the free layer. Writing using pure spin current may improve reliability and data retention compared to STT-MRAM, but the high current used to generate the spin current may lead to design issues related to heating, high power consumption, large transistor size for switching large currents, and the like.

Disclosure of Invention

An apparatus for a magnetoresistive random access memory is presented. In one embodiment, a magnetic tunnel junction for storing data includes a fixed layer, a barrier layer, and a composite free layer. In a certain embodiment, a barrier layer is disposed between the fixed layer and the composite free layer. In another embodiment, the composite free layer includes a ferromagnetic amorphous layer and an in-plane anisotropic free layer. In a certain embodiment, an in-plane anisotropic free layer may be disposed between a ferromagnetic amorphous layer and a barrier layer. In certain implementations, a Spin Hall Effect (SHE) layer can be coupled to a composite free layer of a magnetic tunnel junction. In further embodiments, the SHE layer may be configured such that in-plane currents within the SHE layer apply a torque to the composite free layer, thereby generating spin currents in the composite free layer.

A system for a magnetoresistive random access memory is presented. In one embodiment, a magnetoresistive random access memory die includes a plurality of magnetic tunnel junctions. In a certain embodiment, a magnetic tunnel junction includes a reference layer, a barrier layer, a composite free layer, and a SHE layer including platinum. In another embodiment, a barrier layer is disposed between the reference layer and the composite free layer. In a certain embodiment, a composite free layer is disposed between the SHE layer and the barrier layer. In one embodiment, the composite free layer includes an in-plane anisotropic free layer and a ferromagnetic amorphous layer. In a certain implementation, the in-plane anisotropic free layer may be in contact with a barrier layer. In another embodiment, the ferromagnetic amorphous layer can be in contact with the in-plane anisotropic free layer.

In another embodiment, an apparatus includes means for storing data in a compound free layer for a magnetic tunnel junction based on an orientation of an in-plane magnetic moment of the compound free layer. In a certain embodiment, the apparatus includes means for generating a spin current to change an orientation of an in-plane magnetic moment of the recombination free layer. In some embodiments, the means for generating a spin current comprises platinum. In another embodiment, the device comprises means for separating the means for storing data and the means for generating spin current such that the crystal structure of the means for storing data is not affected by the means for generating spin current.

Drawings

More particular description is now included with reference to specific embodiments that are illustrated in the accompanying drawings. Understanding that these drawings depict only certain embodiments of the disclosure and are not therefore to be considered to be limiting of its scope, the disclosure will be described and explained with additional specificity and detail through the use of the accompanying drawings in which:

FIG. 1 is a schematic block diagram of one embodiment of a system including a Magnetoresistive Random Access Memory (MRAM);

FIG. 2 is a schematic block diagram illustrating one embodiment of an MRAM die;

FIG. 3 is a schematic diagram illustrating one embodiment of a magnetic tunnel junction array;

FIG. 4A is a schematic block diagram illustrating one embodiment of a magnetic tunnel junction including a composite free layer;

FIG. 4B is a schematic block diagram illustrating another embodiment of a magnetic tunnel junction including a composite free layer;

FIG. 5 is a schematic block diagram illustrating one embodiment of a superlattice for a Perpendicular Magnetic Anisotropy (PMA) inducing layer;

FIG. 6 is a graph showing the effective magnetization of an embodiment of the composite free layer in relation to a superlattice structure for the PMA inducing layer;

FIG. 7 is a graph illustrating tunnel magnetoresistance for an embodiment of a magnetic tunnel junction;

FIG. 8 is a schematic flow chart diagram illustrating one embodiment of a method for fabricating a magnetic tunnel junction; and

FIG. 9 is a schematic flow chart diagram illustrating another embodiment of a method for fabricating a magnetic tunnel junction.

Detailed Description

Aspects of the present disclosure may be embodied as an apparatus, system, method, or computer program product. Accordingly, aspects of the present disclosure may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc.) or an embodiment combining software and hardware aspects that may all generally be referred to herein as a "circuit," module, "" device, "or" system. Furthermore, aspects of the present disclosure may take the form of a computer program product embodied in one or more non-transitory computer-readable storage media storing computer-readable program code and/or computer-executable program code.

Many of the functional units described in this specification have been labeled as modules, in order to more particularly emphasize their implementation independence. For example, a module may be implemented as a hardware circuit comprising custom VLSI circuits or gate arrays, off-the-shelf semiconductors such as logic chips, transistors, or other discrete components. A module may also be implemented in programmable hardware devices such as field programmable gate arrays, programmable array logic components, programmable logic devices or the like.

Modules may also be implemented, at least in part, in software for execution by various types of processors. An identified module of executable code may, for instance, comprise one or more physical or logical blocks of computer instructions which may, for instance, be organized as an object, procedure, or function. Nevertheless, the executables of an identified module need not be physically located together, but may comprise disparate instructions stored in different locations which, when joined logically together, comprise the module and achieve the stated purpose for the module.

Indeed, a module of executable code may comprise a single instruction, or many instructions, and may even be distributed over several different code segments, among different programs, across several memory devices, and so forth. When the modules or portions of modules are implemented in software, these software portions may be stored on one or more computer-readable and/or computer-executable storage media. Any combination of one or more computer-readable storage media may be utilized. For example, a computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing, but does not include a propagated signal. In the context of this document, a computer-readable storage medium and/or computer-executable storage medium may be any tangible and/or non-transitory medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, processor, or device.

Computer program code for carrying out operations for aspects of the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as Python, Java, Smalltalk, C + +, C #, Objective C, etc., a conventional procedural programming language such as the "C" programming language, a scripting programming language, and/or other similar programming languages. The program code may execute partially or completely on one or more users' computers and/or on remote computers or servers over a data network or the like.

As used herein, a component includes a tangible, physical, non-transitory device. For example, the components may be implemented as: hardware logic circuits including custom VLSI circuits, gate arrays, or other integrated circuits; off-the-shelf semiconductors such as logic chips, transistors, or other discrete devices; and/or other mechanical or electrical devices. A component may also be implemented in programmable hardware devices such as field programmable gate arrays, programmable array logic components, programmable logic devices or the like. The components may include one or more silicon integrated circuit devices (e.g., chips, dies, die planes, packages) or other discrete electrical devices in electrical communication with one or more other components via wires or the like of a Printed Circuit Board (PCB). In certain embodiments, each of the modules described herein may alternatively be embodied or realized as a component.

As used herein, an electrical circuit includes a set of one or more electrical and/or electronic components that provide one or more paths for electrical current. In certain embodiments, the circuit may include a return path for current such that the circuit is a closed loop. However, in another embodiment, a group of components that do not include a return path for current may be referred to as a circuit (e.g., an open loop). For example, an integrated circuit may be referred to as a circuit whether or not it is grounded (as a return path for current). In various embodiments, the circuit may comprise a portion of an integrated circuit, a set of integrated circuits, a set of non-integrated electrical and/or electronic components with or without integrated circuit devices, and/or the like. In one embodiment, a circuit may include: custom VLSI circuits, gate arrays, logic circuits, or other integrated circuits; off-the-shelf semiconductors such as logic chips, transistors, or other discrete devices; and/or other mechanical or electrical devices. The circuit may also be implemented as a synthesis circuit in a programmable hardware device such as a field programmable gate array, a programmable array logic component, a programmable logic device, or the like (e.g., as firmware, a netlist, or the like). The circuitry may include one or more silicon integrated circuit devices (e.g., chips, dies, die planes, packages) or other discrete electrical devices in electrical communication with one or more other components via wires or the like of a Printed Circuit Board (PCB). In certain embodiments, each of the modules described herein may be embodied or implemented as circuitry.

Reference throughout this specification to "one embodiment," "an embodiment," or similar language means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present disclosure. Thus, appearances of the phrases "in one embodiment," "in an embodiment," and similar language throughout this specification may, but do not necessarily, all refer to the same embodiment, but mean "one or more but not all embodiments" unless expressly specified otherwise. The terms "comprising," "including," "having," and variations thereof mean "including, but not limited to," unless expressly specified otherwise. The enumerated listing of items does not imply that any or all of the items are mutually exclusive and/or mutually inclusive, unless expressly specified otherwise. The terms "a", "an" and "the" also refer to "one or more", unless expressly specified otherwise.

Aspects of the present disclosure are described below with reference to schematic flow charts and/or schematic block diagrams of methods, apparatus, systems, and computer program products according to embodiments of the disclosure. It will be understood that each block of the schematic flow chart diagrams and/or schematic block diagrams, and combinations of blocks in the schematic flow chart diagrams and/or schematic block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a computer or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor or other programmable data processing apparatus, create means for implementing the functions and/or acts specified in the schematic flow chart diagram and/or schematic block diagram block or blocks.

It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. Other steps and methods may be conceived that are equivalent in function, logic, or effect to one or more blocks, or portions thereof, of the illustrated figures. Although various arrow types and line types may be employed in the flow chart diagrams and/or block diagram block diagrams, they are understood not to limit the scope of the corresponding embodiments. For instance, an arrow may indicate a waiting or monitoring period of unspecified duration between enumerated steps of the depicted embodiment.

In the following detailed description, reference is made to the accompanying drawings, which form a part hereof. The foregoing summary is illustrative only and is not intended to be in any way limiting. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features will become apparent by reference to the drawings and the following detailed description. The description of the elements in each figure may refer to elements of previous figures. Like numbers may refer to like elements in the drawings, including alternative embodiments of like elements.

Fig. 1 depicts a system 100 that includes a Magnetoresistive Random Access Memory (MRAM) 150. In the depicted embodiment, the system includes a computing device 110. In various embodiments, computing device 110 may refer to any electronic device capable of computing by performing arithmetic or logical operations on electronic data. For example, the computing device 110 may be a server, a workstation, a desktop computer, a laptop computer, a tablet computer, a smartphone, a control system for another electronic device, an attached network storage device, a block device on a storage area network, a router, a network switch, and so forth. In certain embodiments, the computing device 110 may include a non-transitory computer-readable storage medium storing computer-readable instructions configured to cause the computing device 110 to perform the steps of one or more methods disclosed herein.

In the depicted embodiment, computing device 110 includes a processor 115, a memory 130, and a storage 140. In various implementations, processor 115 may refer to any electronic element that performs arithmetic or logical operations performed by a computing device. For example, in one embodiment, the processor 115 may be a general purpose processor that executes stored program code. In another embodiment, processor 115 may be a Field Programmable Gate Array (FPGA), Application Specific Integrated Circuit (ASIC), or the like, that operates on data stored by memory 130 and/or storage device 140. In a certain embodiment, the processor 115 may be a controller for a storage device (e.g., on a storage area network), a network device, and the like.

In the depicted embodiment, processor 115 includes a cache 120. In various embodiments, the cache 120 may store data used by the processor 115. In some embodiments, cache 120 may be smaller and faster than memory 130, and may replicate data in commonly used locations of memory 130, and the like. In certain embodiments, processor 115 may include multiple caches 120. In various embodiments, cache 120 may include one or more types of memory media for storing data, such as Static Random Access Memory (SRAM)122, Magnetoresistive Random Access Memory (MRAM)150, and so forth. For example, in one embodiment, cache 120 may include SRAM 122. In another embodiment, cache 120 may include MRAM 150. In a certain embodiment, cache 120 may include a combination of SRAM 122, MRAM 150, and/or other memory media types.

In one embodiment, the memory 130 is coupled to the processor 115 by a memory bus 135. In certain embodiments, the memory 130 may store data that is directly addressable by the processor 115. In various embodiments, memory 130 may include one or more types of memory media for storing data, such as Dynamic Random Access Memory (DRAM)132, MRAM 150, and so forth. For example, in one embodiment, memory 130 may include DRAM 132. In another embodiment, memory 130 may comprise MRAM 150. In a certain embodiment, memory 130 may include a combination of DRAM 132, MRAM 150, and/or other memory media types.

In one embodiment, the storage device 140 is coupled to the processor 115 through a storage bus 145. In certain embodiments, the storage bus 145 may be a peripheral bus of the computing device 110, such AS a peripheral component interconnect Express (PCI Express or PCIe) bus, a Serial Advanced Technology Attachment (SATA) bus, a Parallel Advanced Technology Attachment (PATA) bus, a Small Computer System Interface (SCSI) bus, a FireWire bus, a fibre channel connection, a Universal Serial Bus (USB), a PCIe advanced switching (PCIe-AS) bus, and so forth. In various embodiments, storage device 140 may store data that is not directly addressable by processors 115, but is accessible through one or more storage controllers. In certain implementations, storage device 140 may be larger than memory 130. In various embodiments, storage device 140 may include one or more types of storage media for storing data, such as a hard disk drive, NAND flash memory 142, MRAM 150, and so forth. For example, in one embodiment, storage device 140 may include NAND flash memory 142. In another embodiment, storage device 140 may comprise MRAM 150. In a certain embodiment, storage device 140 may include a combination of NAND flash memory 142, MRAM 150, and/or other storage media types.

In various embodiments, the MRAM 150 may be used to store data in the cache 120, the memory 130, the storage 140, and/or another component that stores data. For example, in the depicted embodiment, computing device 110 includes MRAM 150 in cache 120, memory 130, and storage 140. In another embodiment, the computing device 110 may use the MRAM 150 for the memory 130, and may use other types of memory or storage media for the cache 120 or the storage 140. Rather, in another embodiment, computing device 110 may use MRAM 150 for storage 140 and may use other types of storage media for cache 120 and memory 130. Additionally, some types of computing devices 110 may include memory 130 without storage 140 (e.g., in a microcontroller) where memory 130 is non-volatile, may include memory 130 without cache 120 for a special purpose processor 115, and so on.

Various combinations of cache 120, memory 130, and/or storage 140 and uses of MRAM 150 for cache 120, memory 130, storage 140, and/or other applications will be apparent in view of this disclosure.

In various embodiments, MRAM 150 may comprise one or more chips, packages, dies, or other integrated circuit devices comprising magnetoresistive memory disposed on one or more printed circuit boards, memory enclosures, and/or other mechanical and/or electrical support structures. For example, one or more dual in-line memory modules (DIMMs), one or more expansion and/or daughter cards, a Solid State Drive (SSD) or other storage device, and/or another memory and/or storage form factor may include the MRAM 150. The MRAM 150 may be integrated with and/or installed on a motherboard of the computing device 110, installed in a port and/or slot of the computing device 110, installed on a dedicated storage device on a different computing device 110 and/or network, communicate with the computing device 110 over an external bus, and so forth.

In various embodiments, MRAM 150 may include one or more MRAM dies that include multiple Magnetic Tunnel Junctions (MTJs) for storing data. In some implementations, an MTJ includes a fixed layer, a barrier layer, and a composite free layer. The composite free layer may include an in-plane anisotropic free layer, a ferromagnetic amorphous layer, and a Perpendicular Magnetic Anisotropy (PMA) inducing layer. In certain embodiments, the composite free layer including the in-plane anisotropic free layer, the ferromagnetic amorphous layer, and the PMA-inducing layer may reduce write current and power consumption while providing a Tunneling Magnetoresistance (TMR) ratio (e.g., a measure of the difference between a high resistance state and a low resistance state and an anti-parallel state) suitable for reading stored data, as compared to an MTJ with a non-composite free layer. The MRAM 150 is described in more detail below with reference to fig. 2-9.

Fig. 2 depicts one embodiment of MRAM die 150. MRAM die 150 may be substantially similar to MRAM 150 described with reference to fig. 1. In the depicted embodiment, the MRAM die 150 includes an array of magnetic tunnel junctions 200, row circuitry 202, column circuitry 204, and a die controller 206.

In various embodiments, MRAM die 150 may refer to an integrated circuit that includes both a core array 200 (e.g., MTJ) of MRAM cells for magnetoresistive data storage and peripheral components (e.g., row circuitry 202, column circuitry 204, and/or die controller 206) for communicating with array 200. In some embodiments, one or more MRAM dies 150 may be included in a memory module, storage device, or the like.

In the depicted embodiment, the array 200 includes a plurality of magnetic tunnel junctions for storing data. In one embodiment, the array 200 may be a two-dimensional array. In another embodiment, the array 200 may be a three-dimensional array comprising multiple planes and/or layers of MTJs. In various implementations, the array 200 can be addressable by rows (e.g., word lines) by row circuitry 202 and by columns (e.g., bit lines) by column circuitry 204.

In some embodiments, the die controller 206 cooperates with the row circuitry 202 and the column circuitry 204 to perform memory operations on the array 200. In various embodiments, the die controller 206 may include components such as power control circuitry that controls the power and voltages provided to the row and column circuitry 202, 204 during memory operations, address decoders that convert received addresses into hardware addresses used by the row and column circuitry 202, 204, and a state machine that implements and controls memory operations. Die controller 206 may communicate with computing device 110, processor 115, bus controller, memory device controller, memory module controller, etc. via lines 208 to receive command and address information, transmit data, etc.

FIG. 3 depicts one embodiment of a magnetic tunnel junction array 200. The MTJ array 200 may be substantially similar to the MTJ array 200 described with reference to FIG. 2. In the depicted implementation, the MTJ array 200 includes a plurality of MTJs 350 for storing data coupled to a write word line (WL-W)302, a read word line (WL-R)306, and a bit line 304.

In the depicted implementation, the MTJ350 includes a fixed layer or a reference layer (indicated by a single arrow) having a fixed or pinned magnetic moment. In another implementation, the MTJ350 includes a free layer (indicated by the double arrow) having a magnetic moment that can be changed or switched. A thin dielectric or barrier layer may separate the fixed and free layers, and current may flow through the barrier layer due to quantum tunneling. The probability of electron tunneling through the barrier layer is higher if the magnetic moments of the fixed and free layers are substantially parallel to each other (referred to herein as the parallel state of the MTJ 350), and lower if the magnetic moments of the fixed and free layers are substantially anti-parallel to each other (referred to herein as the anti-parallel state of the MTJ 350). Thus, the resistance through the MTJ350 may be higher in the anti-parallel state than in the parallel state.

In various implementations, the difference in resistance between the parallel and anti-parallel states of the MTJ350 allows data to be stored. For example, a low resistance may correspond to a binary "1" and a high resistance may correspond to a binary "0". Alternatively, a low resistance may correspond to a binary "0" and a high resistance may correspond to a binary "1". The difference between the higher anti-parallel resistance and the lower parallel resistance (expressed as a percentage of the lower parallel resistance) may be referred to herein as the Tunnel Magnetoresistance (TMR) or TMR ratio of the MTJ 350. Thus, for example, a TMR of 100% would indicate an anti-parallel resistance that is twice the parallel resistance (e.g., 100% greater than the parallel resistance).

In some implementations, the TMR ratio of the MTJ350 can be related to the difficulty of reading data from the MTJ 350. For example, if the ratio between the anti-parallel resistance and the parallel resistance of the MTJ350 is high, a lower read voltage may be sufficient to produce detectably different read currents in the anti-parallel and parallel states. Conversely, if the ratio between the anti-parallel resistance and the parallel resistance of the MTJ350 is low, the read currents in the anti-parallel and parallel states can be undetectable different read currents unless a higher read voltage is applied. Thus, increasing the TMR of the MTJ350 (or avoiding factors that may reduce the TMR) may facilitate the read operation of the MTJ 350.

In the depicted implementation, the MTJ350 is a spin-orbit torque (SOT) MTJ for which data can be written by applying a current through a Spin Hall Effect (SHE) material adjacent to the free layer, thus generating a pure spin current that is used to apply a magnetic torque to the magnetic moment of the free layer, thereby changing the magnetic moment of the free layer. In various embodiments, SHE material may refer to any material that exhibits a spin hall effect, which induces a spin current perpendicular to the current in the material. For example, in various embodiments, SHE material may include platinum, tungsten, iridium-doped copper, iridium-doped silver, and the like. In certain embodiments, an in-plane current within the SHE material layer may generate a spin current above the layer. The direction of the spin current may be controlled based on the direction of the current. Thus, in certain implementations in which the MTJ350 includes a layer of SHE material adjacent to or in contact with the free layer, an in-plane current within the SHE layer may inject a spin current into the free layer (e.g., by applying a magnetic torque to the free layer) to change the magnetic moment of the free layer.

In the depicted implementation, the write word line 302 and the bit line 304 are coupled to opposite sides of the SHE layer of the SOT-MTJ 350 such that a voltage difference between the write word line 302 and the bit line 304 induces an in-plane current in the SHE layer of the MTJ350 at the intersection of the write word line 302 and the bit line 304. Thus, writing data to the MTJ350 at the intersection of the selected write word line 302 and the selected bit line 304 may include applying a programming voltage to the selected write word line 302 and applying a zero (or other reference) voltage to the selected bit line 304, such that a current in the SHE layer of the MTJ350 applies a torque, thereby generating a spin current in the free layer to change the magnetic moment of the free layer.

In various implementations, the MTJ array 200 can include transistors, selectors, etc. that prevent stray currents through the unselected MTJs 350 during write operations. For example, in one implementation, the voltages of the unselected write word lines 302 and unselected bit lines 304 may be set to half the programming voltage, such that there is no voltage difference between the write word lines 302 and bit lines 304 for the unselected MTJs 350 (e.g., MTJs 350 not on the selected write word line 302 or selected bit line 304), and such that there is a voltage difference of half the programming voltage between the write word lines 302 and bit lines 304 for the half-selected MTJs 350 (e.g., MTJs 350 on either, but not both, the selected write word line 302 or the selected bit line 304). In another implementation, the selector of the MTJ may not allow current to flow unless the voltage difference between the write word line 302 and the bit line 304 is greater than half of the programming voltage. Thus, a selector, transistor, or other switching device may allow a write operation to change stored data in one MTJ350 without changing stored data values in other MTJs 350.

In another implementation, data can be written to a row of MTJs 350 simultaneously. For example, the write word line 302 for a row may be grounded, and a positive or negative programming voltage may be applied to the bit line 304 to write a different data value to the MTJ350 in the row. In view of the present disclosure, various ways of writing data to the single or multiple MTJs 350 in MTJ arrays 200 using various geometries will be apparent.

In various implementations, reading data from the MTJ350 can include measuring, detecting, or sensing the resistance of the MTJ350 (e.g., indicating whether the MTJ350 is in a parallel state or an anti-parallel state). For example, in one embodiment, a known voltage may be applied over the free layer, barrier layer, and fixed layer, and the resulting current may be measured or sensed to detect resistance. In another embodiment, a known current can be applied through the free layer, barrier layer, and fixed layer, and the resulting voltage drop over the MTJ350 can be measured or sensed to detect the resistance. In some embodiments, the MTJ array 200 or the MRAM die 150 may include sense amplifiers, latches, etc. to convert low power signals from the bit line 304 to logic levels representing 1 or 0 and to store the converted data.

In the depicted implementation, the read word line 306 and the bit line 304 are coupled to the MTJ350 to apply a current through the free layer, the barrier layer, and the fixed layer. For example, reading data from the MTJ350 at the intersection of the selected read word line 306 and the selected bit line 304 can include applying a read voltage to the selected read word line 306 and a zero (or other reference) voltage to the selected bit line 304 so that the resistance through the MTJ350 can be sensed. As described above for the write operation, selectors, transistors, and the like can prevent stray currents through the unselected MTJs 350 during the read operation. Additionally, data can be read from a row of MTJs 350 simultaneously. For example, a read voltage may be applied to the read word line 306 of a row and the bit line 304 may be grounded such that the resulting current is sensed for multiple bit lines 304. In view of the present disclosure, various ways of reading data from a single or multiple MTJs 350 in MTJ arrays 200 using various geometries will be apparent.

FIG. 4A depicts one embodiment of a magnetic tunnel junction 350 including a composite free layer 400. The MTJ350 may be substantially similar to the MTJ350 described with reference to FIG. 3. In the depicted implementation, the MTJ350 includes a fixed or reference layer 412, a free layer 400, and a barrier layer 410 disposed between the fixed and free layers 400, which may be substantially as described above with reference to fig. 3. In the depicted embodiment, the free layer 400 is a composite free layer that includes an in-plane anisotropic free layer 408, a Perpendicular Magnetic Anisotropy (PMA) inducing layer 404, and a ferromagnetic amorphous layer 406. Additionally, in the depicted embodiment, the MTJ also includes a spin hall effect layer 402, which may be substantially as described above with reference to fig. 3. In various implementations, the layers of the MTJ350 can be formed or deposited by various techniques such as physical vapor deposition, sputtering, and the like. In some embodiments, additional layers not shown in FIG. 4A, such as capping layers, may be included in the MTJ350, or in the process of fabricating the MTJ 350.

In one embodiment, the fixed or reference layer 412 comprises a ferromagnetic material having a fixed or pinned magnetic moment. As used herein, the term "ferromagnetic" may be used to refer to any material that is capable of spontaneous magnetization (e.g., remains magnetized in the absence of an applied magnetic field). Thus, a "ferromagnetic" material may refer to a strictly ferromagnetic material (e.g., individual microscopic magnetic moments are fully aligned) or a ferrimagnetic material (e.g., individual microscopic magnetic moments are partially anti-aligned).

In various embodiments, a "fixed" or "pinned" magnetic moment refers to a magnetic moment that is at least substantially constant in orientation when the magnetic moment of the free layer 400 is changed or flipped. Thus, for example, in one embodiment, the fixed layer 412 may comprise a ferromagnetic material having a higher coercivity than the ferromagnetic material of the free layer 400. In such implementations, an external magnetic field can change the magnetization of both the fixed layer 412 and the free layer 400, but has a greater effect on the free layer 400. In another embodiment, the fixed layer 412 may comprise a ferromagnetic thin film having a magnetic moment pinned by exchange coupling with an antiferromagnet. For example, in one embodiment, the pinned layer 412 may include a synthetic antiferromagnet (e.g., cobalt/iron and ruthenium multilayer), a ruthenium or iridium spacer, and a ferromagnetic layer including a cobalt/iron/boron alloy (CoFeB).

In various implementations, the magnetic moment of the fixed layer 412 can provide a reference for the orientation of the magnetic moment of the free layer 400. For example, in various implementations, the total magnetic moment of the free layer 400 may be parallel or anti-parallel to the magnetic moment of the fixed layer 412. Thus, the fixed layer 412 is depicted with a reference magnetic moment indicated by a single arrow, and a parallel or anti-parallel magnetic moment of the free layer 400 is indicated by a double arrow.

In various implementations, the barrier layer 410 is disposed between the fixed or reference layer 412 and the free layer 400. In some implementations, barrier layer 410 includes a dielectric material, such as magnesium oxide (MgO). In certain implementations, the barrier layer 410 can be less than 20 angstroms thick such that quantum tunneling of electrons across the barrier layer 410 allows current to flow through the MTJ 350.

In general, in various implementations, the free layer 400 can include a ferromagnetic material having a magnetic moment that can be changed, switched, or flipped relative to the magnetic moment of the fixed layer 412. As described with reference to FIG. 3, changing the magnetic moment of the free layer 400 changes the resistance of the MTJ350, allowing data to be stored. In certain embodiments, the ferromagnetic material of the free layer 400 may comprise a CoFeB alloy. In the depicted embodiment, the free layer 400 is a composite free layer, where the components are described in more detail below. In some implementations, reading the data can include applying a current from terminal T2424 (or terminal T1422) to terminal T3426 to sense the resistance of the MTJ 350.

In one embodiment, the spin hall effect layer 402 comprises a Spin Hall Effect (SHE) material as described above with reference to fig. 3, such as platinum, tungsten, iridium-doped copper, iridium-doped silver, or the like. In various implementations, the SHE layer 402 may be described as part of the MTJ350, or it may be described as coupled to the MTJ (where the MTJ itself includes the fixed layer 412, the free layer 400, and the barrier layer 410). In the depicted embodiment, the SHE layer 402 is configured such that in-plane currents within the SHE layer 402 apply a torque, thereby generating spin currents in the composite free layer 400. For example, in one embodiment, a current from terminal T1422 to terminal T2424 may inject a spin current having a first orientation into the composite free layer 400. In another embodiment, an opposite current from terminal T2424 to terminal T1422 may inject a spin current having an orientation opposite to the first orientation into the composite free layer 400. The spin current can change or flip the magnetic moment of the free layer 400 to write data.

In certain implementations, the current density in the SHE layer 402 for switching the magnetic moment of the non-composite free layer may be on the order of ten million to one hundred million amperes per square centimeter. High switching current densities can lead to design problems related to heating, high power consumption, large transistor dimensions for switching large currents, and the like. However, the switching current density of the free layer 400 having an in-plane magnetic moment may be proportional to the product of the damping constant of the free layer 400, the magnetic thickness of the free layer 400, which is defined as the saturation magnetization multiplied by the thickness, and/or the effective magnetization of the free layer 400, which is defined as the saturation magnetization minus the perpendicular magnetic anisotropy, etc. Thus, in certain implementations, increasing the perpendicular anisotropy of the free layer 400 may reduce the switching current density of the MTJ 350. In various embodiments, the composite free layer 400 may be configured to have a general in-plane anisotropy having a perpendicular component of the anisotropy to reduce the switching current of the MTJ 350. In various implementations, reducing the damping constant, magnetic thickness, and/or effective magnetization of the free layer 400 can reduce the switching current of the MTJ 350. However, reducing the magnetic thickness can make the parallel or anti-parallel state of the MTJ350 less stable and adversely affect data retention. Thus, in certain embodiments, the composite free layer 400 may be configured to reduce switching current by reducing the damping constant and/or effective magnetization of the free layer 400 (relative to the non-composite free layer 400).

In the depicted embodiment, the composite free layer 400 includes an in-plane anisotropic free layer 408, a ferromagnetic amorphous layer 406, and a Perpendicular Magnetic Anisotropy (PMA) inducing layer 404. In various implementations, the in-plane anisotropic free layer 408 may include a ferromagnetic material with a variable in-plane magnetic moment. For example, in one embodiment, the in-plane anisotropic free layer 408 may comprise a ferromagnetic CoFeB alloy. The magnetic moment of the in-plane anisotropic free layer 408 can be changed by spin current from the SHE layer 402, as described above. In certain implementations, the in-plane anisotropic free layer 408 can have a thickness of up to 30 angstroms. In further implementations, the in-plane anisotropic free layer 408 can have a thickness of at least 5 angstroms.

As used herein, terms such as "in-plane" and "perpendicular" may be used to describe a direction or orientation (e.g., a direction or orientation for a vector such as a magnetic moment, magnetization, current density, etc.) relative to the layers of the MTJ 350. In one embodiment, the term "perpendicular" refers to a direction at right angles to the surface of the layer (e.g., the perpendicular direction in fig. 4A), and the term "in-plane" refers to a direction parallel to the surface of the layer (e.g., the horizontal direction in fig. 4A). However, in another embodiment, a vector, orientation, or direction may include a combination of perpendicular and in-plane components, but may be described as "perpendicular" or "in-plane" based on one having a greater magnitude (perpendicular or in-plane component). For example, in one embodiment, although the magnetic moment includes a non-zero in-plane component and a perpendicular component, if the in-plane component is greater than the perpendicular component, that direction or orientation may be described as an "in-plane" magnetic moment.

In various embodiments, magnetic anisotropy may refer to a direction or axis in which magnetization is energetically favorable. For example, in one embodiment, magnetic anisotropy may refer to the direction of the "easy axis" used to magnetize a ferromagnetic material, where the magnetic moment of the ferromagnetic material tends to be oriented along the "easy axis," but may point in either direction along that axis.

In certain embodiments, the in-plane anisotropic free layer 408 has in-plane magnetic anisotropy. In further implementations, the in-plane magnetic anisotropy of the in-plane anisotropic free layer 408 may be parallel or substantially parallel to the magnetic moment of the fixed layer 412. Thus, the magnetic moment of the in-plane anisotropic free layer 408 can be changed by the spin current from the SHE layer 402, but the magnetic moment aligned with the in-plane magnetic anisotropy will be substantially parallel or substantially anti-parallel to the magnetic moment of the fixed layer 412. Thus, in various implementations, the in-plane anisotropic free layer 408 may store data in the composite free layer 400 based on the parallel or anti-parallel orientation of the in-plane magnetic moments.

In one embodiment, the PMA inducing layer 404 is configured to induce, increase or add a perpendicular component to the magnetic anisotropy of the composite free layer 400. In certain embodiments, the PMA inducing layer 404 may permanently add or induce PMA of the composite free layer 400. As described above, the switching current density for writing the MTJ350 can be proportional to the effective magnetization of the free layer 400, where the effective magnetization is defined as the saturation magnetization minus the perpendicular magnetic anisotropy. Thus, in certain implementations, increasing the perpendicular anisotropy may decrease the effective magnetization of the composite free layer 400, thereby decreasing the switching current density of the MTJ 350.

In various embodiments, the PMA inducing layer 404 may comprise any of a variety of materials that induce PMA. For example, in one embodiment, PMA inducing layer 404 may comprise a superlattice of alternating "X" and "Y" layers. In one embodiment, the "X" layer may include 0.5 to 10 angstroms of cobalt and/or iron, and the "Y" layer may include 0.5 to 10 angstroms of platinum, palladium, and/or nickel. The multiple repetitions of the alternating [ X/Y ] structure may be selected to reduce the effective magnetization of the composite free layer 400 by inducing PMA while keeping the total magnetic moment of the composite free layer 400 in-plane. In a certain embodiment, for example, PMA inducing layer 404 may comprise a superlattice of alternating cobalt and platinum layers. In another embodiment, the superlattice of alternating cobalt and platinum layers may include three cobalt layers and three platinum layers. In some embodiments, the cobalt layer may be 3.4 angstroms thick and the platinum layer may be 1.4 angstroms thick.

In another embodiment, the PMA inducing layer 404 may comprise rare earth and transition metal alloys. For example, in one embodiment, the PMA inducing layer 404 may comprise an alloy of cobalt and/or iron (transition metal) having ten to thirty atomic percent gadolinium (rare earth metal). In another embodiment, the PMA inducing layer 404 may comprise L1₀A phase alloy. In certain embodiments, L1 for alloys₀A phase may refer to a structure similar to a face centered cubic crystal structure, but arranged with different elements of the alloy in alternating planes. In one embodiment, L1 for PMA inducing layer 404₀The phase alloy may include an isoatomic alloy of an "X" element and a "Y" element, where the "X" element may be cobalt or iron, and the "Y" element may be platinum, palladium, or nickel. In a certain embodiment, L1₀The phase alloy may be deposited at a temperature greater than (or equal to) 300 degrees celsius. In certain embodiments, the PMA inducing layer 404 may comprise a combination of sub-layers comprising a superlattice, a rare earth and transition metal alloy, L1₀Phase alloys and/or additional PMA induced structures.

In various embodiments, the amount of PMA added by the PMA induction layer 404 to the composite free layer 400 may be proportional to the thickness of the PMA induction layer 404 (or may increase as the thickness of the PMA induction layer increases). Because the effective magnetization of the composite free layer 400 is equal to the saturation magnetization minus the perpendicular anisotropy, increasing the PMA from zero first decreases the effective magnetization of the composite free layer 400, allowing a corresponding decrease in switching current. However, as the total magnetic moment of the composite free layer 400 becomes out-of-plane, further increase of the PMA above the saturation magnetization may increase the effective magnetization and switching current. Thus, in certain embodiments, the thickness of the PMA inducing layer 404 is selected such that the PMA inducing layer 404 reduces the effective magnetization of the composite free layer 400 and such that the overall magnetic moment of the composite free layer 400 is in-plane. In various implementations, the total magnetic moment of the composite free layer 400 may be referred to as in-plane if the in-plane component of the total magnetic moment is greater than the perpendicular component of the total magnetic moment.

In various embodiments, the manufacturer of MTJ350 can control one or more individual alloy layers such as rare earth and transition metal alloys, or L1₀Deposition of the phase alloy and/or by controlling the number of repeating layers of the superlattice structure, the thickness of the PMA inducing layer 404 is controlled. The "reduction" in effective magnetization, switching current density, or switching current may be compared to a composite free layer 400 structure that omits the PMA inducing layer 404.

In the depicted embodiment, the PMA induction layer 404 is disposed such that the in-plane anisotropic free layer 408 is located between the barrier layer 410 and the PMA induction layer 404. For example, the MTJ350 may be constructed with a fixed layer 412 deposited on a substrate, and the in-plane anisotropic free layer 408 may be located above the barrier layer 410 and below the PMA inducing layer 404. In another implementation, the MTJ350 may be constructed with the SHE layer 402 deposited on the substrate, and the in-plane anisotropic free layer 408 may be located below the barrier layer 410 and above the PMA inducing layer 404.

In some implementations, the crystallization of the in-plane anisotropic free layer 408 may proceed from the interface with the barrier layer 410. For example, in one embodiment, the in-plane anisotropic free layer 408 may be deposited as an amorphous structure and may be annealed such that the in-plane anisotropic free layer 408 and the barrier layer 410 crystallize in a pseudo-epitaxial relationship. Matched or similar crystal structures (e.g., pseudo-epitaxy) in the fixed layer 412, barrier layer 410, and in-plane anisotropic free layer 408 may facilitate quantum tunneling through the barrier layer 410, and TMR for such structures may be suitable for reading data. In contrast, a mismatch in crystal orientation within the in-plane anisotropic free layer 408, at an interface between the in-plane anisotropic free layer 408 and the barrier layer 410, or at another interface between a layer and a layer may impede quantum tunneling, such that the resistance of the MTJ350 is high in the parallel and anti-parallel states, and the low TMR for such structures may make the MTJ350 less suitable or unsuitable for storing data.

In certain implementations, the crystal structure of the PMA inducing layer 404 and/or the SHE layer 402 may be different from the crystal structure of the barrier layer 410. For example, in various embodiments, the barrier layer 410 may have a cubic or body centered cubic crystal structure, and the PMA inducing layer 404 and/or the SHE layer 402 may have a face centered cubic crystal structure. In another embodiment, if the PMA inducing layer 404 and/or SHE layer 402 is to directly contact the in-plane anisotropic free layer 408, crystallization from both sides of the in-plane anisotropic free layer 408 during annealing may cause the crystals within the in-plane anisotropic free layer 408 to be mismatched, resulting in a low TMR for the MTJ 350.

Thus, in certain embodiments, the ferromagnetic amorphous layer 406 is disposed between the in-plane anisotropic free layer 408 and the PMA inducing layer 404. For example, in one embodiment, the in-plane anisotropic free layer 408 may be in contact with the barrier layer 410, the ferromagnetic amorphous layer 406 may be in contact with the in-plane anisotropic free layer 408, and the PMA inducing layer 404 may be in contact with the ferromagnetic amorphous layer 406.

In certain embodiments, the ferromagnetic amorphous layer 406 can comprise any material that is both ferromagnetic and amorphous. In certain implementations, the ferromagnetic materials of the PMA induction layer 404 and the ferromagnetic amorphous layer 406 may conduct spin current from the SHE layer 402 into the in-plane anisotropic free layer 408. In contrast, non-ferromagnetic materials (such as heavy metal atoms) can scatter spin current, thus impeding the write operation to the MTJ 350.

In certain embodiments, the amorphous structure of the ferromagnetic amorphous layer 406 allows the crystallization of the in-plane anisotropic free layer 408 during annealing to be based on the crystalline structure of the barrier layer 410, rather than on the crystalline structure of the PMA inducing layer 404. For example, the ferromagnetic amorphous layer 406 may separate the in-plane anisotropic free layer 408 and the PMA induction layer 404 such that the crystalline structure of the in-plane anisotropic free layer 408 is not affected by the PMA induction layer 404. Thus, if the PMA inducing layer 404 or another layer other than the barrier layer 410 that is not amorphous directly contacts the in-plane anisotropic free layer 408, the ferromagnetic amorphous layer 406 can prevent or mitigate the TMR reduction of the MTJ350 that may otherwise occur.

In one embodiment, the ferromagnetic amorphous layer 406 can include an alloy of one or more ferromagnetic elements, one or more glass-forming elements, and one or more stabilizing elements for preventing migration of the one or more glass-forming elements. (as used herein, when referring to an alloy, "element" specifically refers to the type of atom, rather than more generally to the component.) the ferromagnetic element may be iron, cobalt, nickel, or any other element that exhibits ferromagnetism. In one embodiment, the one or more ferromagnetic elements may include iron, cobalt, or a combination of iron and cobalt.

In various embodiments, a glass-forming element can refer to any element that tends to hinder or prevent the crystallization of one or more ferromagnetic elements, such that the ferromagnetic amorphous layer 406 remains in an amorphous solid or glassy state. For example, in a certain embodiment, the one or more glass-forming elements may include boron. However, in some embodiments, the glass-forming elements may migrate within the ferromagnetic amorphous layer 406 at temperatures typical for annealing the in-plane anisotropic free layer 408, allowing crystallization to begin in regions with a low concentration of glass-forming elements such that the ferromagnetic amorphous layer 406 is no longer amorphous. Thus, in certain embodiments, the ferromagnetic amorphous layer 406 can include one or more stabilizing elements for preventing migration of one or more glass-forming elements. (however, in another embodiment, the one or more glass-forming elements may not migrate sufficiently to allow crystallization during annealing, and the one or more stabilizing elements may be omitted.)

In various embodiments, the stabilizing element can comprise any element that tends to reduce or prevent migration of the glass-forming element in the ferromagnetic amorphous layer 406. For example, the high chemical affinity between boron and tantalum, titanium, or zirconium may prevent boron migration within the ferromagnetic amorphous layer 406, thereby maintaining the amorphous structure. In one embodiment, the one or more stabilizing elements may include tantalum, titanium, or a combination of tantalum and titanium.

In one embodiment, the ferromagnetic amorphous layer 406 can include an alloy having ten atomic percent or more of the one or more glass-forming elements. In a certain embodiment, the alloy may include five atomic percent or less of the one or more stabilizing elements. In some embodiments, the alloy may include at least three percent of the one or more stabilizing elements.

In one embodiment, the ferromagnetic amorphous layer 406 includes an alloy of cobalt, iron, and/or nickel having at least ten atomic percent boron and at least three atomic percent tantalum, titanium, and/or zirconium. In another embodiment, the ferromagnetic amorphous layer 406 includes an alloy of cobalt, iron, and/or nickel having at least twenty atomic percent hafnium, zirconium, silicon, germanium, tantalum, and/or niobium.

In one embodiment, the ferromagnetic amorphous layer 406 comprises an alloy of cobalt, titanium, and boron (CoTiB). In another embodiment, the ferromagnetic amorphous layer 406 comprises an alloy of cobalt, iron, boron, and tantalum (CoFeBTa). In certain embodiments, the CoTiB alloy and CoFeBTa alloy may remain amorphous even after annealing at temperatures up to 700K.

FIG. 4B depicts another embodiment of a magnetic tunnel junction 350 including a composite free layer 400. The MTJ350 may be substantially similar to the MTJ350 described above with reference to FIG. 3 and FIG. 4A. In the depicted implementation, the MTJ350 includes a fixed or reference layer 412, a free layer 400, and a barrier layer 410 disposed between the fixed and free layers 400, which may be substantially as described above with reference to fig. 3 and 4A. In the depicted embodiment, the free layer 400 is a composite free layer that includes an in-plane anisotropic free layer 408 and a ferromagnetic amorphous layer 406, which may be substantially as described above with reference to FIG. 4A. Additionally, in the depicted embodiment, the MTJ also includes a spin hall effect layer 402, which may be substantially as described above with reference to fig. 3 and 4A.

In the depicted implementation, the MTJ350 does not include the PMA inducing layer 404. As shown in FIG. 4A, the in-plane anisotropic free layer 408 is disposed between the ferromagnetic amorphous layer 406 and the barrier layer 410. However, in fig. 4B, the ferromagnetic amorphous layer 406 is disposed in contact with the SHE layer 402 without the PMA inducing layer 404.

In some implementations, barrier layer 410 may include a material having a body-centered cubic crystal structure, such as magnesium oxide. In another embodiment, the fixed layer 412 and/or the in-plane anisotropic free layer 408 may be deposited as an amorphous material and crystal formation during annealing may proceed from the interface with the barrier layer 410 such that the fixed layer 412 and/or the in-plane anisotropic free layer 408 has (or includes) a body-centered cubic crystal structure. In contrast, the SHE layer 402 and/or the PMA inducing layer 404 described above may comprise a material having a face centered cubic structure, such as platinum.

Thus, in various embodiments, the ferromagnetic amorphous layer 406 may be disposed between the in-plane anisotropic free layer 408 and the PMA inducing layer 404 (as depicted in fig. 4A) or between the in-plane anisotropic free layer 408 and the SHE layer 402 (as depicted in fig. 4B). In certain embodiments, disposing the ferromagnetic amorphous layer 406 may break the crystalline continuity between the body-centered cubic material of the in-plane anisotropic free layer 408 and the face-centered cubic material of the PMA inducing layer 404 or the SHE layer 402. If the SHE layer 402 or another layer other than the barrier layer 410 that is not amorphous directly contacts the in-plane anisotropic free layer 408, the ferromagnetic amorphous layer 406 may prevent or mitigate the TMR reduction of the MTJ350 that may otherwise occur by avoiding a mismatched interface between face-centered cubic material and body-centered cubic material. Additionally, in certain embodiments, the ferromagnetic amorphous layer 406 may reduce the damping constant of the free layer 400 and/or may reduce the effective magnetization of the free layer 400, thus reducing the switching current (as compared to the free layer 400 without the ferromagnetic amorphous layer 406).

Fig. 5 depicts one embodiment of a superlattice of PMA inducing layers 404. In various embodiments, a superlattice may refer to periodic layered structures. In the depicted embodiment, the PMA induction layer 404 comprises a superlattice of alternating cobalt layers 502 and platinum layers 504. In various implementations, the thickness of the separate cobalt layer 502 and platinum layer 504 may be less than 10 angstroms. In one embodiment, cobalt layer 502 may be 3.4 angstroms thick and platinum layer 504 may be 1.4 angstroms thick. Multiple repetitions of the cobalt layer 502 and the platinum layer 504 may be selected or controlled during fabrication to reduce the effective magnetization of the composite free layer 400 (by inducing the PMA) while keeping the overall magnetic moment of the composite free layer 400 in-plane.

Fig. 6 is a graph depicting the effective magnetization of an embodiment of the composite free layer 400 in relation to multiple repetitions of alternating cobalt and platinum layers (e.g., cobalt layer 502 and platinum layer 504 of fig. 5) of the superlattice of PMA inducing layer 404. The effective magnetization is normalized in fig. 6 such that the effective magnetization with zero repetition (e.g., without PMA inducing layer 404) is equal to 1. The graph shows that the effective magnetization is reduced with additional repetitions so that the effective magnetization with three cobalt layers and three platinum layers is between the normalized values 0.1 and 0.2. Thus, in certain embodiments, the PMA inducing layer 404 may reduce the effective magnetization of the composite free layer 400 by one fifth to one tenth.

In another implementation, the switching current of the MTJ may be reduced in proportion to the reduction in the effective magnetization of the composite free layer 400. For example, in one embodiment, the switching current density may be up to 5 million amperes per square centimeter without the PMA inducing layer 404, and the PMA inducing layer 404 may reduce the switching current density by an order of magnitude to five million amperes per square centimeter.

In a certain embodiment, the in-plane current within the SHE layer 402 applies a torque to the composite free layer 400, thereby generating a spin current in the composite free layer 400, and the PMA inducing layer 404 reduces the current density of the in-plane current to between one million amperes per square centimeter and ten million amperes per million amperes. In another embodiment, the PMA inducing layer 404 may reduce the current density of the in-plane current to between three and seven million amperes per square centimeter. In a certain embodiment, the PMA inducing layer 404 may reduce the current density of the in-plane current to between four and six million amperes per square centimeter.

However, excessive repetition of alternating cobalt and platinum layers of the superlattice of the PMA inducing layer 404 may induce sufficient PMA such that the magnetic moment of the composite free layer 400 is no longer in-plane. For example, the decreasing trend in the graph of fig. 6 indicates that four or more repetitions of the superlattice structure may cause the effective magnetization of the composite free layer 400 to be negative or out-of-plane.

FIG. 7 is a graph depicting Tunnel Magnetoresistance (TMR) for an embodiment of a magnetic tunnel junction. As described above, TMR may refer to the difference between a higher anti-parallel resistance and a lower parallel resistance of the MTJ, expressed as a percentage of the lower parallel resistance. In various embodiments, higher TMR corresponds to a more pronounced difference between the anti-parallel and parallel states, and increasing TMR may be beneficial in reducing read current and/or voltage. The graph in fig. 7 depicts TMR in the absence of the PMA inducing layer 404, such that the crystal structure of the in-plane anisotropic free layer 408 is affected by the SHE layer 402 and/or the ferromagnetic amorphous layer 406.

In the depicted graph, the TMR of an MTJ with the tantalum SHE layer 402 and without the ferromagnetic amorphous layer 406 is higher than 125%. In contrast, the TMR of an MTJ with the platinum SHE layer 402 and without the ferromagnetic amorphous layer 406 is lower than 100% and close to 75%. The face centered cubic structure of platinum in the SHE layer 402 and the body centered cubic structure of magnesium oxide in the barrier layer 410 may mis-crystallize the in-plane anisotropic free layer 408 during annealing, resulting in a reduction in TMR. However, the introduction of a 10 angstrom thick CoTiB ferromagnetic amorphous layer 406 restored the TMR with the platinum SHE layer 402 to about 100%. Therefore, the TMR reduction due to the mis-crystallization can be sufficiently prevented or mitigated by the ferromagnetic amorphous layer 406.

FIG. 8 is a schematic flow chart diagram illustrating one embodiment of a method 800 for fabricating a magnetic tunnel junction. The method 800 begins and the manufacturer deposits 802 a fixed layer 412 on the substrate. The manufacturer deposits 804 barrier layer 410 on fixed layer 412. The manufacturer deposits 806 the in-plane anisotropic free layer 408 on the barrier layer 410. The manufacturer deposits 808 a ferromagnetic amorphous layer 406 on the in-plane anisotropic free layer 408. The manufacturer deposits 810 the PMA inducing layer 404 or superlattice onto the ferromagnetic amorphous layer 406. The manufacturer deposits 812 the spin hall effect layer 402 on the PMA induction layer 404 and the method 800 ends.

FIG. 9 is a schematic flow chart diagram illustrating another embodiment of a method 900 for fabricating a magnetic tunnel junction. The method 900 begins and the manufacturer deposits 902 the spin hall effect layer 402 on the substrate. The manufacturer deposits 904 the PMA inducing layer 404 on the spin hall effect layer 402. The manufacturer deposits 906 a ferromagnetic amorphous layer 406 on the PMA inducing layer 404. The manufacturer deposits 908 the in-plane anisotropic free layer 408 on the ferromagnetic amorphous layer 406. The manufacturer deposits 910 the barrier layer 410 on the in-plane anisotropic free layer 408. The manufacturer deposits 912 fixed layer 412 on barrier layer 410 and method 900 ends.

In various implementations, the means for storing data in the composite free layer 400 may include an in-plane anisotropic free layer 408, a ferromagnetic material, a ferromagnetic alloy, a CoFeB alloy, or the like. Other embodiments may include similar or equivalent means for storing data in the composite free layer 400.

In various embodiments, means for permanently inducing Perpendicular Magnetic Anisotropy (PMA) for the composite free layer 400 may include a PMA inducing layer 404, a superlattice, a rare earth and transition metal alloy, L1₀Phase alloys, and the like. Other embodiments may include similar or equivalent means for inducing PMA in the composite free layer 400.

In various embodiments, the means for separating the means for storing data and the means for generating spin current may comprise a ferromagnetic amorphous layer 406, one or more ferromagnetic elements, one or more glass-forming elements, one or more stabilizing elements for preventing migration of the one or more glass-forming elements, a CoTiB alloy, a CoFeBTa alloy, or the like. Other embodiments may include similar or equivalent means for separating the means for storing data from the means for generating spin currents.

In various implementations, the means for generating a spin current to change the orientation of the in-plane magnetic moment of the composite free layer 400 may include the SHE layer 402, a tantalum layer, a platinum layer, a write word line, a bit line, a word line driver, a power supply, and the like. Other embodiments may include similar or equivalent means for generating spin currents.

The present disclosure may be embodied in other specific forms without departing from its spirit or essential characteristics. The described embodiments are to be considered in all respects only as illustrative and not restrictive. The scope of the disclosure is, therefore, indicated by the appended claims rather than by the foregoing description. All changes which come within the meaning and range of equivalency of the claims are to be embraced within their scope.

Claims

1. An apparatus, comprising:

a magnetic tunnel junction for storing data, the magnetic tunnel junction comprising a fixed layer, a barrier layer, and a composite free layer, the barrier layer disposed between the fixed layer and the composite free layer, the composite free layer comprising:

a ferromagnetic amorphous layer; and

an in-plane anisotropic free layer disposed between the ferromagnetic amorphous layer and the barrier layer; and

a Spin Hall Effect (SHE) layer coupled to the composite free layer of the magnetic tunnel junction, the SHE layer configured such that an in-plane current within the SHE layer applies a torque to the composite free layer, thereby generating a spin current in the composite free layer.

2. The apparatus of claim 1, wherein the ferromagnetic amorphous layer comprises an alloy of one or more ferromagnetic elements, one or more glass-forming elements, and one or more stabilizing elements for preventing migration of the one or more glass-forming elements.

3. The device of claim 2, wherein the one or more ferromagnetic elements comprise one or more of iron and cobalt, the one or more glass-forming elements comprise boron, and the one or more stabilizing elements comprise one or more of tantalum and titanium.

4. The apparatus of claim 2, wherein the alloy comprises ten atomic percent or more of the one or more glass-forming elements and five atomic percent or less of the one or more stabilizing elements.

5. The apparatus of claim 1, wherein the ferromagnetic amorphous layer comprises an alloy of cobalt, titanium, and boron.

6. The apparatus of claim 1, wherein the ferromagnetic amorphous layer comprises an alloy of cobalt, iron, boron, and tantalum.

7. The apparatus of claim 1, wherein the composite free layer further comprises a Perpendicular Magnetic Anisotropy (PMA) inducing layer disposed between the SHE layer and the ferromagnetic amorphous layer, a thickness of the PMA inducing layer being selected such that the PMA inducing layer reduces an effective magnetization of the composite free layer and such that an overall magnetic moment of the composite free layer is in-plane.

8. The apparatus of claim 7, wherein the PMA inducing layer decreases the effective magnetization of the composite free layer by one fifth to one tenth.

9. The apparatus of claim 7, wherein the PMA inducing layer comprises one or more of: superlattice of alternating cobalt and platinum layers, rare earth and transition metal alloys, and L1₀A phase alloy.

10. The apparatus of claim 7, wherein the PMA inducing layer comprises a superlattice of alternating cobalt and platinum layers, the superlattice comprising three cobalt and three platinum layers.

11. The apparatus of claim 1, wherein the SHE layer comprises platinum.

12. The apparatus of claim 1, wherein the SHE layer comprises a face centered cubic crystal structure, the in-plane anisotropic free layer comprises a body centered cubic crystal structure, and the ferromagnetic amorphous layer is disposed between the SHE layer and the in-plane anisotropic free layer.

13. A system, comprising:

a Magnetoresistive Random Access Memory (MRAM) die comprising a plurality of magnetic tunnel junctions, wherein a magnetic tunnel junction comprises a reference layer, a barrier layer, a composite free layer, and a Spin Hall Effect (SHE) layer comprising platinum, the barrier layer disposed between the reference layer and the composite free layer, the composite free layer disposed between the SHE layer and the barrier layer, the composite free layer comprising:

an in-plane anisotropic free layer in contact with the barrier layer; and

a ferromagnetic amorphous layer in contact with the in-plane anisotropic free layer.

14. The system of claim 13, wherein the ferromagnetic amorphous layer comprises an alloy of one or more ferromagnetic elements, one or more glass-forming elements, and one or more stabilizing elements for preventing migration of the one or more glass-forming elements.

15. The system of claim 14, wherein the one or more ferromagnetic elements comprise one or more of iron and cobalt, the one or more glass-forming elements comprise boron, and the one or more stabilizing elements comprise one or more of tantalum and titanium.

16. The system of claim 14, wherein the alloy comprises ten atomic percent or more of the one or more glass-forming elements and five atomic percent or less of the one or more stabilizing elements.

17. The system of claim 13, wherein the composite free layer further comprises a Perpendicular Magnetic Anisotropy (PMA) inducing layer disposed between the SHE layer and the ferromagnetic amorphous layer, a thickness of the PMA inducing layer being selected such that the PMA inducing layer reduces an effective magnetization of the composite free layer and such that an overall magnetic moment of the composite free layer is in-plane.

18. The system of claim 13, wherein the SHE layer is configured such that in-plane currents within the SHE layer generate spin currents in the composite free layer.

19. An apparatus, comprising:

means for storing data in a composite free layer for a magnetic tunnel junction based on an orientation of an in-plane magnetic moment of the composite free layer;

means for generating a spin current to change the orientation of the in-plane magnetic moment in the composite free layer, wherein the means for generating a spin current comprises platinum; and

means for separating the means for storing data and the means for generating spin current such that a crystal structure of the means for storing data is not affected by the means for generating spin current.

20. The apparatus of claim 19, further comprising means for permanently inducing a Perpendicular Magnetic Anisotropy (PMA) for the composite free layer.