A nanoscale spin rectifier device

Abstract

This document describes a nanoscale spin rectifier device for harvesting ambient radiofrequency (RF) energy comprising a reference magnetic layer, a free magnetic layer and a spacer layer disposed between the reference magnetic layer and the free magnetic layer. The nanoscale spin rectifier device may also be implemented in a spin rectifier rectenna system for harvesting ambient RF energy. Additionally, a plurality of such nanoscale spin rectifier devices may be connected in a series and / or parallel configuration to form an ambient RF energy harvesting module, with output terminals coupled to an electrical load to receive the rectified voltage.

Description

A NANOSCALE SPIN RECTIFIER DEVICECROSS REFERENCE TO RELATED APPLICATION

[0001] This application claims the benefit of priority to Singapore patent application no. 10202401955R which was filed on 3 July 2024, the contents of which are hereby incorporated by reference in its entirety for all purposes.TECHNICAL FIELD

[0002] This application relates to a nanoscale spin rectifier device for harvesting ambient radiofrequency (RF) energy. The device includes a reference magnetic layer, a free magnetic layer and a spacer layer disposed between the reference magnetic layer and the free magnetic layer. The nanoscale spin rectifier device may also be configured for use in a spin rectifierbased rectenna system to harvest ambient RF energy. Additionally, multiple nanoscale spin rectifier devices may be connected in a series configuration to form an ambient RF energy harvesting module, with output terminals optionally coupled to an electrical load.BACKGROUND

[0003] The digital revolution has brought about transformative changes in everyday life, with wireless sensor networks (WSNs) becoming essential in applications such as health monitoring and environmental sensing. As these networks continue to proliferate under the broader framework of the Internet of Things (loT), a critical challenge persists, the elimination of the dependency of digital devices on batteries to ensure long operational lifetimes in compact and often inaccessible locations. Sensors deployed in difficult-to-access locations for monitoring variables like air quality, temperature, and humidity must operate autonomously and ideally without the need for battery replacement or maintenance.

[0004] To support such battery-free operation, one approach proposed by those skilled in the art is to harness ambient radiofrequency (RF) energy, which is continuously emitted by the vast infrastructure of wireless communication systems. As billions of loT devices communicate and exchange data globally, a significant portion of the ambient RF energy in the environment remains untapped. Those skilled in the art have theorized that if this energy could be effectively harvested and converted into direct-current (DC) power, it could provide a sustainable energy source for low-power electronic sensors and devices. Among the various energy harvesting methods available, such as solar, thermal, vibration, and wind, RF energyharvesting stands out for its continuous availability, ease of integration, and suitability for small-form-factor WSN deployments.

[0005] A standard RF energy harvesting module (EHM) typically consists of a receiving antenna, a rectifier to convert the incoming RF signal into DC power, a power management circuit, and a load that consumes the harvested power. For ambient-RF harvesting applications, the EHM must operate efficiently at low RF power levels, and these levels typically range from -82 dBm to -20 dBm, which correspond to approximately 6.3 picowatts to 10 microwatts. These levels are characteristic of the ambient RF spectrum near the widely used 2.4 GHz band. Among the components used in the EHM, the rectifier is a key component that determines overall power conversion efficiency. Rectifiers employed by those skilled in the art usually rely on Schottky diodes, transistors, or CMOS technologies, with Schottky diodes that are implemented on both rigid and flexible substrates remaining the most efficient choice for RF-to-DC conversion.

[0006] Despite their relatively high efficiency, existing rectifier technologies face notable limitations when operating under ultra-low power conditions (i.e., when the RF power is below -20 dBm). This is due to fundamental thermodynamic constraints and the need for supporting components such as matching circuits and external antennas, which hinder their miniaturization and on-chip integration. While some implementations using Schottky diodes have demonstrated moderate efficiencies at RF power levels between -30 and -20 dBm, their performance remains inadequate for practical deployment in energy harvesting modules targeting lower power regimes. As such, those skilled in the art are constantly seeking for rectification technologies that are able to offers high sensitivity to ambient RF energy below -20 dBm, while also supporting ultra-compact, chip-scale integration for next generation WSN and loT applications.SUMMARY

[0007] In one aspect, the present application discloses a nanoscale spin rectifier device for harvesting ambient radiofrequency energy. The disclosed device comprises a reference magnetic layer having a fixed magnetization direction, a spacer layer disposed on the reference magnetic layer, the spacer layer configured to enable spin dependent tunnelling of electrons between the reference layer and a free magnetic layer disposed on the spacer layer. Inembodiments of the disclosure, the free magnetic layer has an equilibrium magnetization direction that is canted relative to the fixed magnetization direction of the reference magnetic layer, the equilibrium magnetization direction being determined by a thickness of the free magnetic layer and a thickness of the spacer layer. Additionally, RF signals incident on the device induces precessional motion of the magnetization direction of the free magnetic layer which modulates a tunnelling resistance across the spacer layer to generate a rectified DC voltage across the device.

[0008] In embodiments of the one aspect, the generation of the rectified DC voltage occurs without requiring an external direct-current (DC) bias or magnetic field to be applied across the device. Tn further embodiments, the device further comprises a synthetic antiferromagnetic structure disposed below the reference magnetic layer and an exchange bias layer disposed below the synthetic antiferromagnetic structure, the exchange bias layer formed from an antiferromagnetic material, wherein the reference magnetic layer and the free magnetic layer each comprise a cobalt-iron-boron (CoFeB) alloy, and the synthetic antiferromagnetic structure comprises a multilayer magnetic stack.

[0009] In embodiments of the one aspect, the multilayer magnetic stack of the synthetic antiferromagnetic structure comprises a cobalt (Co) layer, a ruthenium (Ru) layer, and a CoFeB layer, wherein the synthetic antiferromagnetic structure is exchange-biased by an antiferromagnetic material layer comprising platinum-manganese (PtMn), iron-manganese (FeMn), or iridium-manganese (IrMn), and the spacer layer comprises magnesium oxide (MgO).

[0010] In embodiments of the one aspect, the free magnetic layer comprises a thickness between 1 nm and 6 nm, the spacer layer comprises a thickness between 0.6 nm and 3 nm, and the device comprises lateral dimensions between 40 x 50 nm2and 80 x 200 nm2.

[0011] In embodiments of the one aspect, an ambient radiofrequency energy harvesting module is disclosed whereby the module comprises a plurality of the devices according to the embodiments described above. In this embodiment, the plurality of devices are connected in a series, parallel or hybrid-series-parallel configuration and have a pair of output terminals coupled to an electrical load. In further embodiments, the plurality of devices comprises atleast ten nanoscale spin rectifier devices. In still further embodiments, the plurality of the devices comprise between two and ten nanoscale spin rectifier devices connected in the series configuration, wherein a total rectified DC voltage across the electrical load increases with a number of the plurality of devices connected in series. In still yet further embodiments, the plurality of the devices comprise between two and ten nanoscale spin rectifier devices connected in the parallel configuration, wherein a total rectified DC across the electrical load increases with a number of the plurality of devices connected in parallel.

[0012] In another aspect, the present application discloses a method for forming a nanoscale spin rectifier device for harvesting ambient radiofrequency energy, the method comprising the steps of forming a reference magnetic layer having a fixed magnetization direction; forming a spacer layer on the reference magnetic layer; and forming a free magnetic layer on the spacer layer. In embodiments of the disclosure, the spacer layer is configured to enable spin dependent tunnelling of electrons between the reference layer, and the free magnetic layer has an equilibrium magnetization direction that is canted relative to the fixed magnetization direction of the reference magnetic layer, the equilibrium magnetization direction being determined by a thickness of the free magnetic layer and a thickness of the spacer layer. Additionally, RF signals incident on the device induces precessional motion of the magnetization direction of the free magnetic layer which modulates a tunnelling resistance across the spacer layer to generate a rectified DC voltage across the device.

[0013] In embodiments of the another aspect, a method for providing a spin rectifier rectenna system for harvesting ambient radiofrequency energy is disclosed. The disclosed method comprises the step of electrically coupling a high-gain antenna to the device formed according to previous embodiment, wherein the high-gain antenna is configured to be impedance matched to an operating frequency of the device.

[0014] In embodiments of the another aspect, a method for providing an ambient radiofrequency energy harvesting module is disclosed. The disclosed method comprises the steps of connecting a plurality of the devices according to the embodiments above in a series, parallel or hybrid-series-parallel configuration, and coupling a pair of output terminals of the module to an electrical load.BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Various embodiments of the present disclosure are described below with reference to the following drawings:Figure 1 illustrates a block diagram of energy harvesting modules powering electronic components using ambient radiofrequency energy as is known in the art;Figure 2 illustrates a schematic of an energy harvesting module in accordance with embodiments of the disclosure;Figure 3 illustrates a front view of an embodiment of a nanoscale spin rectifier device in accordance with embodiments of the present disclosure;Figure 4 illustrates a front view of another nanoscale spin rectifier device according to one embodiment of the present disclosure;Figure 5a illustrates a frontal view of an embodiment of an elliptical shaped nanoscale spin rectifier device in accordance with embodiments of the present disclosure;Figure 5b illustrates a frontal view of another elliptical shaped nanoscale spin rectifier device according to one embodiment of the present disclosure;Figure 6 illustrates experimentally measured rectified voltages (Vr) of 40 x 100 nm2and 80 x 200 nm2nanoscale spin rectifier devices as functions of the resonant frequency of the incident RF signal under zero-bias and zero magnetic field conditions;Figure 7 illustrates experimentally measured sensitivity responses for a 40 * 100 nm2spin rectifier device at an operating frequency of 3.5 GHz and for an 80 x 200 nm2spin rectifier device at an operating frequency of 2.45 GHz as a function of the power of the incident RF signal;Figure 8a illustrates a schematic of a spin rectifier antenna system in accordance with embodiments of the disclosure;Figure 8b illustrates an exemplary design of a patch antenna designed to match an impedance of the nanoscale spin rectifier device in accordance with embodiments of the disclosure;Figure 8c illustrates experimentally measured rectified voltages (Vr) of an 80 x 200 nm2spin rectifier device when the device is attached to a matched antenna or a 50 Ohm antenna, when the power of the incident RF signal is varied;Figure 8d illustrates experimentally measured sensitivity (S) of an 80 x 200 nm2spin rectifier device when the device is attached to a matched antenna or a 50 Ohm antenna, when the power of the incident RF signal is varied;Figure 9a illustrates experimentally measured broadband rectification results for a single 40 x 100 nm2spin rectifier device and an array of ten such spin rectifier devices connected in series, under an incident RF signal with a power level of -20 dBm;Figure 9b illustrates experimentally measured resonant and broadband rectification results for a single 80 x 200 nm2spin rectifier device and an array of ten such spin rectifier devices connected in series, under an incident RF signal with a power level of -20 dBm;Figure 10a illustrates a simulated phase diagram showing the precession amplitude of the x- component of magnetization (Amx), which governs the tunnelling magnetoresistance (TMR) response, as a function of microwave frequency (f) and alternating current density amplitude (Jac);Figure 10b illustrates simulated results of the maximum precession amplitude of the x- component of magnetization (Amx) as a function of alternating current density (Jac) for the 40 x 100 nm2single spin rectifier device;Figure 10c illustrates simulated results of the precession amplitude of the x-component of magnetization (Amx) as a function of microwave frequency (f), showing the transition from resonant to broadband behaviour;Figure lOd illustrates experimentally observed emission spectra from a single spin rectifier (plot 1002) and two spin rectifiers connected in series (plot 1004), both excited by an RF signal under zero direct-current (DC) bias (Ide = 0 mA),Figure lOe illustrates the extracted second harmonic (2f) peak power (Pzr), showing an enhancement when spin rectifiers are connected in series;Figure I la illustrates a schematic of the energy harvesting module, in which the rectified voltage (Vr) generated by the spin rectifier-array is initially stored in a capacitor and subsequently stepped up to a high voltage (Vstcp ~ 1.6 - 4 V) using a boost converter to power a temperature sensor;Figure 1 lb illustrates the peak rectified voltage (Vr) generated by the spin rectifier-array when excited using 2.45 GHz and 3.5 GHz antennas;Figure 11 c illustrates the discharge behaviour of the stored stepped-up voltage Vstepat an input power level of Prf = -25 dBm (from two rf sources), across different electrical loads;Figure 12a illustrates the conversion efficiency, r|(%), of Schottky diodes, spin rectifier-arrays and spin rectifier antennas at 2.45 GHz with varying input power levels Prr;Figure 12b illustrates the sensitivity, (S), of Schottky diodes, spin rectifier-arrays and spin rectifier antennas at 2.45 GHz with varying input power levels Prf,Figure 13a illustrates the tunnelling magnetoresistance (TMR) of the spin rectifier device as a function of the spacer layer thickness, for free magnetic layers having a thickness of 1.7 nm, 1.8 nm and 1.9 nm respectively;Figure 13b illustrates the resonance frequency corresponding to the maximum rectification value of the spin rectifier device as a function of the spacer layer thickness, for free magnetic layers having a thickness of 1.7 nm, 1.8 nm and 1.9 nm respectively;Figure 13c illustrates the RF sensitivity at the resonance frequency of the spin rectifier device as a function of the spacer layer thickness, for free magnetic layers having a thickness of 1.7 nm, 1.8 nm and 1.9 nm respectively;Figure 14a illustrates the magnetoresistance of a 40 x 100 nm2spin rectifier, showing a canted equilibrium magnetization that arises from the competition between in-plane shape anisotropy and out-of-plane interfacial anisotropy;Figure 14b illustrates the magnetoresistance of an 80 x 200 nm2spin rectifier, demonstrating a canted equilibrium magnetization resulting from the competition between in-plane shape anisotropy and out-of-plane interfacial anisotropy;Figure 15a illustrates the tunnelling magnetoresistance (TMR) and the resistance-area (RA) product as a function of the junction dimension of the spin rectifier device;Figure 15b illustrates canted equilibrium angle as a function of the junction dimension of the spin rectifier device;Figure 15c illustrates the zero-bias rectification signal at an input power signal of Prf = -30 dBm as a function of operating frequency, for junction dimensions of 120 x 300 nm2(x 10), 100 x 250 nm2, 80 200 nm2, 60 x 150 nm2, and 40 x 100 nm2;Figure 15d illustrates the frequency of peak rectification voltage and corresponding zero bias RF-sensitivity as a function of the junction dimension of the spin rectifier device;Figure 16a illustrates the shift of the perpendicular anisotropic field within the spin rectifier device as a function of the DC bias voltage, for junction dimensions of 40 x 100 nm2, 60 x 150 nm2, 80 x 200 nm2, and 100 x 250 nm2;Figure 16b illustrates the voltage-controlled magnetic anisotropy (VCMA) coefficient as extracted from the slope of the plots in Figure 16a;Figure 17 illustrates the rectified voltage as a function of varying RF power, for junction dimensions of 40 x 100 nm2and 80 x 200 nm2;Figure 18a illustrates the on-chip impedance of a spin rectifier device based on the reflection coefficient Sn as measured by vector network analyser (VNA);Figure 18b illustrates the reflection coefficient of a patch antenna, impedance matched with a spin-rectifier;Figure 18c illustrates the 3D radiation pattern of a patch antenna, impedance-matched with the spin rectifier (SR), designed using a computer simulation technology;Figure 18d illustrates the received power, as measured by the vector network analyzer (VNA), as a function of the frequency of the incident RF signal when the RF signal is applied from a near-field distance of 2.5 cm, for various input RF power levels;Figure 18e illustrates the received power, as measured by the VNA, as a function of input RF power (Ptf) for various near-field distances;Figure 19a illustrates the measured voltage rectification of a spin rectifier device as a function of the frequency under two magnetic field conditions;Figure 19b illustrates the variation in the peak rectified voltage of the spin rectifier device as a function of the applied magnetic field;Figure 20a illustrates the peak rectified voltage as a function of the number of spin rectifier devices connected in series for junction dimensions of 40 100 nnf and 80 200 nm2;Figure 20b illustrates the rectification bandwidth at which the peak rectification voltage reduces by half as a function of the number of spin rectifier devices connected in series for junction dimensions of 40 x 100 nm2and 80 200 nm2;Figure 21 illustrates the simulated precession amplitude of the x-component of magnetization (Amx) as a function of alternating current density (Jac);Figure 22a illustrates the rectification voltage of a spin rectifier device as a function of time, forRF input signals having various operating frequencies;Figure 22b illustrates the energy harvesting module voltage of a spin rectifier device as a function of time, for RF input signals having various operating frequencies;Figure 23a illustrates the signal-to-noise ratio of various spin rectifier designs as a function of the input RF power;Figure 23b illustrates the noise-equivalent-power of a spin rectifier device as a function of the input RF power;Figure 24 illustrates the efficiency of various spin rectifier designs as a function of the input RF power;Figure 25a illustrates a schematic layout of a co-planer waveguide ground-signal-ground design for use with a spin rectifier device;Figure 25b illustrates a photograph of a co-planer waveguide ground-signal -ground design as implemented on a printed circuit board; andFigure 26 illustrates a flowchart showing a process for forming a spin rectifier device in accordance with embodiments of the disclosure.DETAILED DESCRIPTION

[0016] The following detailed description is made with reference to the accompanying drawings, showing details and embodiments of the present disclosure for the purposes of illustration. Features that are described in the context of an embodiment may correspondingly be applicable to the same or similar features in the other embodiments, even if not explicitly described in these other embodiments. Additions and / or combinations and / or alternatives as described for a feature in the context of an embodiment may correspondingly be applicable to the same or similar feature in the other embodiments.

[0017] In the context of various embodiments, the articles “a”, “an” and “the” as used with regard to a feature or element include a reference to one or more of the features or elements. The term “and / or” includes any and all combinations of one or more of the associated listed items.

[0018] In the context of various embodiments, the term “about” or “approximately” as applied to a numeric value encompasses the exact value and a reasonable variance as generally understood in the relevant technical field, e.g., within 10% of the specified value.

[0019] As used herein, “comprising” means including, but not limited to, whatever follows the word “comprising”. Thus, use of the term “comprising” indicates that the listed elements are required or mandatory, but that other elements are optional and may or may not be present.

[0020] As used herein, “consisting of’ means including, and limited to, whatever follows the phrase “consisting of’. Thus, use of the phrase “consisting of’ indicates that the listed elements are required or mandatory, and that no other elements may be present.

[0021] It should be noted that although the terms first, second and third are used herein to describe various elements, these elements should not be limited by these terms as these terns are meant to only distinguish one element from another element. Thus, the first element described herein could be termed as a second element without departing from this disclosure.

[0022] In the context of various embodiments, the term “surround” means to enclose something completely to form a barrier around it. Thus, the use of the term “surround” indicates that something is on all sides of another thing.

[0023] In the context of various embodiments, the term “disposed on" relates to the placement or deposition of one material or layer onto the surface of another and may involve one or more types of deposition techniques.

[0024] In the context of various embodiments, the term “around” or “adjacent” means to be in the proximity or location of something and does not necessarily mean that two objects have to be in contact.

[0025] In the context of various embodiments, the directional terms mentioned herein, such as “above” and “below” or “upper” and “lower” refer to directions as described with reference to the drawings. Therefore, the directional terms are only used for illustration and are not meant to limit the present disclosure.

[0026] As used herein, a “layer” refers to a material portion including a region having a particular thickness. The layer may extend over the entirety of the structure or may cover only part of the structure as defined in the description. For example, a layer may be located between two horizontal planes; may be located between, or at, a top surface and a bottom surface of the structure. The layer may also extend horizontally, vertically, and / or along the surface of the structure.

[0027] Additionally, for the sake of brevity, extensive explanations of conventional techniques of fabricating semiconductor devices and integrated circuits are not described in detail herein. The tasks and processes described herein may also be integrated into a more comprehensive procedure with extra steps of features that are not elaborated upon in this document. Specifically, certain processes of fabricating semiconductor devices are well known to one skilled in the art hence, such processes will be omitted entirely.

[0028] Radiofrequency (RF) energy harvesting from ambient wireless signals offers a promising route to reduce the carbon footprint of electronic devices A block diagram of energy harvesting modules 102 powering electronic components 104 using ambient radiofrequency energy is illustrated in Figure 1 However, conventional RF rectifiers used in such energy harvesting modules 102 face significant limitations at low ambient power levels (below -20 dBm), primarily due to thermodynamic constraints and parasitic impedance at high frequencies. While nanoscale spin-rectifiers (SRs) based on magnetic tunnel junctions have recently shown high sensitivity, their AC-to-DC conversion efficiency remains low (typically below 1%) and they face challenges such as narrow rectification bandwidths of less than 2 GHz.

[0029] Schottky diodes and tunnel diodes are currently the two leading technologies used for high-efficiency RF rectification. Among these, Schottky diodes are the most widely commercialized, particularly in high-frequency on-chip applications. However, these conventional DC-biased rectifiers are fundamentally limited by thermodynamic constraints, which hinder their performance at low power levels and high frequencies. As a result, they are not well-suited for ambient energy harvesting scenarios, where the incident RF power is typically weak (below -20 dBm).

[0030] In contrast, the spin rectifier (SR) device and an ultra-sensitive spin rectifier connected with an external antenna (SR-rectenna) described herein are capable of harvesting ambient RF signals in the extremely low power range of -62 to -20 dBm. Furthermore, the SR device is able to operate without requiring an additional external antenna, making it highly advantageous for on-chip and battery-free applications.

[0031] This disclosure also describes an on-chip SR-array whereby each SR device in the SR-array is integrated with or without its own compact co-planar waveguide structure. The disclosed SR-array is able to achieve a zero-bias sensitivity of approximately 34,500 mV / mW and a rectification efficiency of up to 7.81%. A key advantage of the SR device disclosed herein lies in the leveraging of self-parametric excitation enabled by voltage-controlled magnetic anisotropy (VCMA), which enhances the magnetization dynamics critical for effective rectification.

[0032] Notably, the SR device and SR-array described in this disclosure is able to operate without the need for external magnetic excitation. Unlike conventional spintronic devices that often rely on applied magnetic fields to induce or sustain magnetization dynamics, the SR device disclosed herein leverages self-parametric excitation driven by VCMA. This mechanism enables efficient precession of the free layer magnetization solely through incident RF signals and voltage-induced anisotropy modulation. As a result, the SR device is able to achieve high-sensitivity and broadband rectification under ambient RF conditions and supports magnet-free operational design.

[0033] A schematic of an energy harvesting module in accordance with embodiments of the disclosure is illustrated in Figure 2. Energy harvesting module 200 comprises antenna 201 that is coupled to rectifier 202. Rectifier 202 is configured to convert RF signals, Prf, received by antenna 201 into rectified DC output voltage, Pdc, via the spin-diode effect. Specifically, upon receiving an RF signal, rectifier 202 induces a spin-transfer torque on the local spins within its free magnetic layer, initiating magnetization precession. This precessional motion leads to time-varying resistance at the same frequency as the incident RF signal. The interaction between this resistance oscillation and the RF current results in the generation of a rectified DC output voltage across the terminals of rectifier 202. In some embodiments, the rectified DC output voltage is supplied to a power management module 204, which regulates or stores the harvested power before delivering it to a load 206.

[0034] Figure 3 illustrates a nanoscale spin rectifier device that may be utilized in energy harvesting module 200 in accordance with embodiments of the present disclosure. Nanoscale spin rectifier device 300 comprises reference magnetic layer 302 that has a fixed magnetization direction. In some embodiments, the fixed magnetization direction may be oriented along an in-plane axis of the device, such as along the longitudinal x-direction of spin rectifier device 300.

[0035] Spacer layer 304 is disposed on reference magnetic layer 302 and free magnetic layer 306 is in turn disposed on spacer layer 304. Spacer layer 304 is configured to enable spin-dependent tunneling of electrons between reference magnetic layer 302 and free magnetic layer 306. Spin-dependent tunneling of electrons refers to a quantum mechanical tunnelling process in which the probability of electron transmission through spacer layer 304depends on the relative orientation of the magnetization directions of the reference 302 and free 306 magnetic layers. In other words, when the magnetization directions of these two layers are aligned in a parallel configuration, the tunnelling probability is higher, resulting in lower electrical resistance. Conversely, when the magnetization directions of these two layers are antiparallel or non-aligned (such as in a canted state), the tunnelling probability decreases, leading to a higher resistance. This variation in resistance with magnetization alignment gives rise to the tunnelling magnetoresistance (TMR) effect, which enables the RF signal rectification effect in spin rectifier device 300.

[0036] In embodiments of the disclosure, free magnetic layer 306 has an equilibrium magnetization direction that is canted relative to the fixed magnetization direction of reference magnetic layer 302 where the equilibrium magnetization direction is determined by a thickness of free magnetic layer 306 and a thickness of spacer layer 304. This canted configuration enhances the spin-diode effect by enabling RF-induced precession of the free layer magnetization, which modulates the tunneling resistance across the spacer layer 304 and results in the generation of a rectified DC voltage. The layered structure may optionally be formed on a substrate (not shown) and includes additional buffer or capping layers such as conductive layers (not shown) as required for structural and functional optimization.

[0037] In accordance with another embodiment of the disclosure, spin rectifier device may incorporate a synthetic antiferromagnetic structure disposed below or formed beneath reference magnetic layer 302 and an exchange bias layer disposed below the synthetic antiferromagnetic structure. Such a device is illustrated in Figure 4. Specifically, device 400 comprises synthetic antiferromagnetic structure 404 that is disposed below reference magnetic layer 302. As shown, exchange bias layer 402 which is formed from an antiferromagnetic material, such as platinum-manganese (PtMn), iron-manganese (FeMn), or iridiummanganese (IrMn), is disposed below synthetic antiferromagnetic structure 404. In this embodiment, synthetic antiferromagnetic structure 404 may include a multilayered magnetic stack comprising a cobalt (Co) layer, a ruthenium (Ru) layer and a CoFeB alloy layer, wherein the synthetic antiferromagnetic structure is exchange-biased by the antiferromagnetic material layer.

[0038] In an exemplary embodiment of the disclosure, device 400 may be formed on a sapphire substrate (not shown) and may comprise exchange bias layer 402 formed from PtsgMita having a thickness of 15 nm, synthetic antiferromagnetic structure 404 formed from a Co layer having a thickness of2.4 nm and a Ru layer having a thickness of 0.88 nm, reference magnetic layer 302 formed from C018.75F 656.25625 having a thickness of 2.4 nm, spacer layer 304 formed from MgO having an optimizable thickness of (tMgo), and free magnetic layer 306 formed from Coi8.75Fe56.2sB25 having an optimizable thickness of (tcoFeB). Conductive layers may also be formed on free magnetic layer 306 and between the substrate and exchange bias layer 402 to facilitate electrical coupling of device 300 or 400.

[0039] The metallic and dielectric layers in devices 300 and 400 may be deposited using DC and RF magnetron sputtering techniques, respectively, under a base pressure of less than 1 x 10'6Pa in an argon atmosphere at room temperature. In embodiments of the disclosure, the thicknesses of free magnetic layer 306 (tcoFeB)) and spacer layer 304 (tMgo) may be systematically varied between 1.7-1.9 nm (or 1 - 6 nm )and 0.9-1.2 nm (or 0.6 - 3 nm), respectively, to tailor the magnetic anisotropy of devices 300 and 400, and to achieve a canted equilibrium magnetization direction of free magnetic layer 306, which is tilted out of plane (z-axis) from the in-plane (x-axis) fixed magnetization direction of reference magnetic layer 302. In embodiments of the disclosure, reference magnetic layer 302 may have its magnetization pinned along the x-axis direction of device 400 via synthetic antiferromagnetic structure 404 comprising a Co and Ru layers, which is exchange-biased by adjacent antiferromagnetic PtMn layer 402.

[0040] The performance of devices 300 and 400 may be optimized by tuning the thicknesses of free magnetic layer 306 (tcoi eu) and spacer layer 304 (tMgo) and it was found that devices having a free magnetic layer 306 (tcoFeB) thickness of 1.9 nm and a spacer layer 304 (tMgo) thickness of 1.0 nm exhibited the most favourable rectification characteristics.

[0041] To further enhance the performance of these devices, the multilayer stack of devices 300 and 400 may be patterned into elliptical nanopillar structures with lateral dimensions ranging from 40 x 100 nm2to 160x400 nm2, using electron beam lithography followed by argon ion beam milling, while maintaining an aspect ratio of 2.5. The elliptical geometry was deliberately selected for these devices as it introduces in-plane shape anisotropy that stabilizesthe magnetization direction of free magnetic layer 306 and enables controlled magnetization dynamics under RF excitation This anisotropic confinement facilitates efficient spin-diode rectification by enhancing the coherence and amplitude of the magnetization precession, thereby improving the rectified output of devices 300 and 400.

[0042] A frontal view of embodiments of elliptical shaped nanoscale spin rectifier devices in accordance with embodiments of the present disclosure are illustrated in Figures 5a and 5b. Figure 5a illustrates spin rectifier device 500 that has been patterned into an elliptical structure, wherein device 500 has a horizontal elliptical cross-sectional shape that has a major axis and a minor axis. In embodiments of the disclosure, the major and minor axes of device 500 may have an aspect ratio of 2.5. As illustrated, device 500 comprises free magnetic layer 306 that is disposed on spacer layer 304 which in turn is disposed on reference magnetic layer 302. Figure 5b illustrates spin rectifier device 502, which has a similar structure as device 500 with the addition of synthetic antiferromagnetic structure 308 that is formed below reference magnetic layer 302 and exchange bias layer 310 that is disposed beneath synthetic antiferromagnetic structure 308. It should be noted that the interactions amongst layers 302- 310 correspond to interactions previously described above in relation to devices 300 and 400.

[0043] In accordance with embodiments of the disclosure, nanoscale spin rectifier devices 300, 400, 500 and 502 may be designed for GHz-range RF rectification under zero-bias and zero magnetic field conditions. In an exemplary embodiment, both reference and free magnetic layers 302 and 306 respectively, may be formed from a cobalt-iron-boron (CoFeB) alloy, with spacer layer 304 comprising a magnesium oxide (MgO) material. To recap, the thicknesses of free magnetic layer 306 and spacer layer 304, along with the lateral dimensions of these devices, may be specifically engineered or optimized to tune the interfacial magnetic anisotropy and TMR of these devices, as these parameters together influence the zero-bias resistance, resonance frequencies, and the equilibrium magnetization direction of free magnetic layer 306 in these devices.

[0044] In embodiments of the disclosure, it was found that SR devices with dimensions of 40 x 100 nm2and 80 x 200 nm2demonstrated superior zero-bias rectification behaviour and this may be attributed to their canted equilibrium magnetization. It was also found that larger devices, such as those with dimensions of 100 * 250 nm2and 160 * 400 nm2, showed reducedrectification efficiency. Hence, for subsequent analysis and demonstration of scalability of these devices, the 40 x 100 nm2and 80 x 200 nm2devices were selected as representative configurations in later sections of this disclosure.

[0045] Further, in embodiments of the disclosure, through systematic variation of these structural parameters, it was determined that devices having free magnetic layer 306 that had a thickness of approximately 1.9 nm, spacer layer 304 that had a thickness of 1 nm, and lateral dimensions ranging between 40 x 50 nm2and 80 x 200 nm2delivered optimal rectification performance. Figure 6 illustrates the rectification response of SR devices with lateral sizes of 40 x l00 nm2and 80 * 200 nm2under zero-bias and zero magnetic field conditions, demonstrating their ability to generate a measurable rectified voltage when subjected to low incident RF power (e g., Prf = -30 dBm).

[0046] Figure 6 shows that the devices exhibited strong broadband rectification responses across frequencies relevant to commercial wireless standards, including WiFi (2.4 GHz), 4G (2 3-2.6 GHz), and 5G (3.5 GHz), under extremely low input RF power levels (e g , Prf = -30 dBm). The rectified voltage spectra of these devices show a combination of symmetric and anti-symmetric Lorentzian components, consistent with spin-diode behaviour previously reported in similar MTJ-based devices.

[0047] Figure 7 illustrates the sensitivity of SR devices with lateral dimensions of 40 x 100 nm2at an operating frequency of 3.5 GHz and a dimension of 80 x 200 nm2at an operating frequency of 2.45 GHz as a function of the input RF power levels As illustrated, it can be seen that the 80 x 200 nm2SR device achieved a maximum zero-bias sensitivity of approximately 2400 mV / mW at an input RF power of -55 dBm, while the 40 x 100 nm2SR device exhibited a peak sensitivity of approximately 1000 mV / mW at -40 dBm. These values exceed the zero-bias sensitivity reported in prior MTJ-based spin rectifier studies as well as that of state-of-the-art Schottky diode rectifiers.

[0048] The superior performance of the 80 x 200 nm2SR device may be attributed to a combination of high zero-field TMR, characterized by Rzb— (Rzb— Rp) / Rpand a strong voltage-controlled magnetic anisotropy (VCMA) where Rzbis defined as the resistance under zero magnetic field, and Rpcorresponds to the resistance in the parallel magnetizationconfiguration. Additionally, it was found that the 80 x 200 nnf device demonstrates a VCMA coefficient exceeding 50 fJ / Vm and a zero-field TMR ratio greater than 30%. Since the rectified output voltage scales with TMR, achieving a Rzb> 30% assists in the generation of a large rectified signal under zero-bias and zero-field conditions. Additionally, a high VCMA enhances device sensitivity by introducing nonlinear contributions to the spin-diode response. It was also found that the VCMA values observed in these devices are comparable or higher than those typically reported for MTJ-based spin rectifiers, which are in the range of 30- 50 fJ / V m.

[0049] Furthermore, it was determined that maintaining a low resistance-area (RA) product is beneficial for the SR device, as high RA values can significantly limit the flow of RF current through the device. An additional design consideration is the inclusion of canted anisotropy, which enhances zero-field TMR and enables strong diode behaviour without requiring external bias or magnetic fields. Accordingly, spin rectifiers exhibiting high ARzb, low RA, and large VCMA coefficients are well suited for efficient ambient RF energy harvesting. It was also noted that at higher input RF power levels (Prf > -20 dBm), the sensitivity of the RF device begins to decline due to reductions in ARzband saturation of the rectified output voltage (Vr), as observed in Figure 7.

[0050] In embodiments of the disclosure, to support the optimization of the SR devices described in the previous sections, RF excitation and rectification measurements may be performed using a probe station setup with the rectification characteristics of the SRs being evaluated via the spin-torque ferromagnetic resonance (ST-FMR) technique. In this method, an RF signal modulated at a low frequency of 213 Hz is supplied by a signal generator, and the resulting rectified DC voltage is measured using a lock-in amplifier configured with a 100 ms time constant and a 24 dB / octave low-pass filter. During the initial measurement step, it was determined that an RF power loss ranging from 35% to 60%, which was measured across the 2.4 to 2.5 GHz frequency range, may be attributed to impedance mismatch between the 50 Q terminated probes and the SR devices. This mismatch is characterized using the reflection coefficient (Si i) as measured by a vector network analyzer (VNA). Notably, this RF loss is not factored into the sensitivity extraction. For probe station-based measurements, the RF input power (Prt refers to the actual signal power delivered from the generator to thedevice through the probes, while in antenna-based configurations, Prf is determined using the transmission coefficient (S21) as measured by the VNA.

[0051] To mitigate the power loss of approximately 35-60% that was observed at 2.4 GHz when an RF signal was injected into the SR device using the probe station or external antenna and to enhance sensitivity, a high-gain, impedance-matched receiving antenna (Rx) may be designed and integrated with a SR device, such as, but not limited to, the 80 x 200 nm2SR device, forming a spin rectifier antenna (SR-rectenna), as shown in Figure 8a. In particular, Figure 8a illustrates antenna 801 that is coupled to an impedance matched circuit 802, which is configured to match the input impedance of SR device or rectifier 202 at an operating frequency of 2.45 GHz. The SR-rectenna is also provided with nanovoltmeter 804 which is electrically connected across the output terminals of SR device or rectifier 202 and is used to measure the rectified DC voltage generated by the device during RF excitation.

[0052] Figure 8b illustrates an exemplary layout of patch antenna 801, which is specifically designed to achieve impedance matching with rectifier 202. The dimension of the transmission line may be optimized to match the SR impedance at 2.45 GHz. As shown, patch antenna 801 is fabricated from a conductive copper layer 806 deposited on a dielectric substrate 808, such as Rogers 4003C. This design ensures efficient RF energy transfer from patch antenna 801 to the rectifier 202 by minimizing impedance mismatch losses.

[0053] As shown in Figures 8c and 8d, the use of the impedance-matched SR-rectenna (as illustrated in Figure 8a) leads to a substantial improvement in both the rectified voltage and sensitivity, particularly under ultra-low input power conditions (Prf < -50 dBm). Specifically, Figure 8c illustrates experimentally measured rectified voltages (Vr) of a 80 * 200 nm2spin rectifier device when the device is attached to a matched antenna or a 50 Ohm antenna, as the incident RF signal power is varied and Figure 8d illustrates experimentally extracted sensitivity (S) of a 80 * 200 nm2spin rectifier device when the device is attached to a matched antenna or a 50 Ohm antenna, when the power of the incident RF signal is varied.

[0054] Figure 8c illustrates that for the impedance-matched SR-rectenna, a rectified voltage of Vr = 6.27 iiV is achieved at an input RF power level, Prf = -62 dBm, and from Figure 8d, it can be seen that this corresponds to a sensitivity of approximately10,000 mV / mW, significantly exceeding the zero-bias sensitivity values of 450- 1850 mV / mW reported in prior SR devices, as well as the 2400 mV / mW sensitivity of the unmatched SR device, i.e. the 50 Ohm antenna. Furthermore, from Figure 8d, it can be seen that the matched SR-rectenna maintains a high sensitivity (S > 1000 mV / mW) across a broad range of input powers from -62 to -25 dBm. Hence, with a favorable signal-to-noise ratio (SNR) of ~20 dB, it can be seen that the matched SR-rectenna can operate reliably at input powers as low as -62 dBm, making it well suited for ultra-low power wireless energy harvesting applications. Figure 8d also demonstrates that the sensitivity achieved using conventional wired excitation with a 50 antenna is comparable to that obtained through wireless excitation with the impedance-matched SR-rectenna. This consistency in performance confirms the successful integration and effectiveness of the SR rectenna architecture for harvesting ambient RF energy under practical operating conditions.

[0055] Although the impedance-matched SR-rectenna exhibits significantly enhanced RF sensitivity, the resulting output voltage remains insufficient to directly power the energy harvesting module (EHM). To overcome this limitation, it is proposed that the output voltage be increased by configuring multiple SR devices in an array. In contrast to traditional Schottky diode cascades, which typically require two to four stages for voltage multiplication, the proposed SR devices can be conveniently interconnected using short wire bonds or predefined electrodes, resulting in a more compact form factor. Furthermore, the inherent compatibility of SRs with CMOS technology suggests strong potential for on-chip integration, offering additional advantages for scalable energy harvesting solutions.

[0056] The measured broadband rectification results for a single 40 x 100 nm2spin rectifier device and an array of ten such spin rectifier devices connected in series, under an incident RF signal with a power level of -20 dBm are illustrated in Figure 9a. As for Figure 9b, this figure illustrates experimentally measured resonant and broadband rectification results for a single 80 x 200 nm2spin rectifier device and an array of ten such spin rectifier devices connected in series, under an incident RF signal with a power level of -20 dBm. It should be noted that the broadband rectification measurements may be conducted using a high-gain (6- 7 dBi), broadband (0.8-18 GHz) horn antenna to efficiently couple RF energy to the SR device and SR array. One skilled in the art will recognize that SR devices with other lateral dimensions may also be used without departing from this disclosure.

[0057] The 40 * 100 nm2SR-device demonstrates a broadband rectification response 904 spanning 0. 1-3.5 GHz at higher input RF powers, i.e., Prf > —25 dBm, as shown in Figure 9a. In contrast, the 80 x 200 nm2SR-device maintained a distinct resonant response under low input RF power conditions Pr< —30 dBm (as illustrated in Figure 6) and a distinct resonant response 908 under high input RF power conditions, i.e., Prf > -25 dBm as illustrated in Figures 6 and 9b, respectively. Additionally, in other embodiments of the disclosure, it was determined that when an applied magnetic field at SR-device is tuned, an extended bandwidth of 0.1-6 GHz may be achieved with a single SR device, which is approximately twice the bandwidth reported in existing SR devices. This will be discussed in detail in the later sections.

[0058] In embodiments of the disclosure, an array of ten SR devices (comprising either device 300 or 400) for both types of 40 x 100 nm2and 80 x 200 nm2SR-devices, were connected in series and directly excited by ambient wireless RF energy without the use of external antennas. This configuration addresses a key challenge in the development of miniaturized EHMs. To minimize on-chip area consumption, a plurality of compact co-planar waveguides (CPWs) may be employed to couple the RF power to each individual SR whereby each individual SR may be coupled to its own CPW. In embodiments of the disclosure, each of the CPWs may have the dimensions of 240 x 500 pm2. One skilled in the art will recognize that although ten SR devices were described in this embodiment, the number of SR devices in the array may comprise more than ten SR devices without departing from this disclosure. In other embodiments of the disclosure, the RF power may be wirelessly coupled directly to each individual SR or an array of SR devices without the need for a CPW to further minimize the on-chip area.

[0059] As shown in Figure 9a, the 40 x l OO nm2SR-array demonstrated a broadband frequency response 902 consistent with that of the individual SR devices when the input RF power Prf > —25 dBm. Notably, the SR-array exhibited a broadband frequency response 902 across a wide frequency range even in the absence of an external magnetic field. This enhancement may be attributed to a stronger self-parametric excitation effect in the SR array, making it well suited for fully integrated, on-chip RF energy harvesting applications.

[0060] A qualitatively different response was observed for the 80 x 200 nm2SR-array, as shown in Figure 9b. Specifically, a transition from resonant behaviour 908 to broadband behaviour 906 was noted, characterized by a narrower detection bandwidth compared to the 40 x l00 nm2array, but with significantly enhanced sensitivity. This suggests that the 80 x 200 nm2SR-array may be particularly advantageous for use as a band-pass filter- integrated rectifier. In terms of scaling, the peak rectification response of the array increased approximately linearly with the number of SRs (up to four or five devices), beyond which the response began to deviate from ideal N-fold linear scaling (i.e., the scaling became less significant) and eventually saturated. Nevertheless, increasing the number of SRs in the series configuration generally resulted in a broadened frequency response and a higher cumulative DC output voltage.

[0061] In another embodiment of the disclosure, a plurality of spin rectifier (SR) devices, such as those having lateral dimensions of 40 x 100 nm2or 80 x 200 nm2, may be connected in a parallel configuration to form the SR-array. In contrast to the series configuration, where the rectified output voltage increases with the number of SR devices, the parallel configuration results in an increase in the total rectified direct current (DC) when the number of SR devices increases. In particular, the total output current of the parallel SR-array scales as with the number of SR devices when low impedance matching is required, and each SR device is individually coupled to its own co-planar waveguide (CPW) for efficient RF energy coupling. This configuration also preserves the broadband or resonant behaviour characteristic of the individual SR devices, depending on the chosen lateral dimensions and operating power regime. Accordingly, a parallel SR-array may serve as a compact, high-current source in on- chip RF energy harvesting modules (EHMs), complementing series configurations optimized for high-voltage applications.

[0062] In yet another embodiment, a hybrid series-parallel configuration may comprise a grouping configuration where each group comprises multiple SR devices connected in series, with each group subsequently connected in parallel, or each group may comprise multiple SR devices connected in parallel, with each group subsequently connected in series, and this configuration may be employed to simultaneously enhance both the rectified DC voltage and current. This architecture leverages the voltage scaling benefit of series-connected SRs, whilethe parallel connection of SRs increases the total output current while maintaining lower impedance.

[0063] Micromagnetic simulations were performed to replicate the experimentally observed transition from resonant to broadband rectification behaviour in a single 40 x 100 nm2SR device. Figure 10a illustrates a phase diagram that maps the rectified voltage response as a function of both frequency and the amplitude of the applied alternating current density Jac clearly illustrating the transition from resonant to broadband behaviour. Figure 10b plots the amplitude variation of the x-component of the magnetization vector, Amx(aligned with the reference layer magnetization), as a function of Jac. Representative rectification curves obtained from the simulations are plotted in Figure 10c where at the lowest input power level (e.g., Jac= 0.1 MA / cm2), the SR device exhibits a narrow resonant response with a small magnetization precession angle. As the input power level increases to Jac= 10 MA / cm2, a pronounced magnetization precession (Amx) emerges, resulting in a broadband rectification response. This transition mechanism is characteristic of smaller SRs with a canted equilibrium magnetization angle between 50° and 70°. In contrast, for larger 80 x 200 nm2SRs, where the canted equilibrium angle approaches the out-of-plane direction (approximately 79° to 95°), the transition from resonant to broadband behaviour is not observed.

[0064] To investigate the potential origin of the transition from resonant to broadband response in the 40 * 100 nm2SR-array, additional experimental studies were conducted. Specifically, microwave emission spectra were recorded using a spectrum analyzer for both a single SR device and a configuration with two SRs connected in series, each driven by an RF current iac= Iacsin (mt) at frequency m. Figure lOd illustrates the experimentally observed emission spectra from the single SR device (spectra 1002) and two SRs connected in series (spectra 1004), both excited by an RF signal under zero direct-current bias (Ide = 0 mA) conditions

[0065] It should be noted that in the spectrum analyzer measurements, the output signal originating from SR oscillations coexists spectrally with the reflected component of the input RF signal (Pri . Although a directional coupler may be employed to isolate the SR-generated signal from the incident RF signal, complete separation is not achieved due to inherent limitations. However, effective distinction between the two is possible owing to theirmarkedly different spectral characteristics, where the SR oscillation signal exhibits a linewidth on the order of a few MHz, while the reflected component of the input RF signal possesses a significantly narrower linewidth of only a few Hz. Given that the spectral resolution of the analyzer is on the order of a few kHz, and substantially broader than the linewidth of the reflected signal, the injected RF signal manifests as a single narrow data point within the spectrum. This narrow component can be reliably removed through appropriate background subtraction using Lorentzian fitting of the overall measured response, thereby isolating the true SR signal.

[0066] As can be seen from spectra 1002 of Figure lOd, the single SR exhibits a weak second harmonic (2<D) emission at 6 GHz. Tn contrast, as can be seen from spectra 1004, the series-connected SRs demonstrate a significant enhancement in the second harmonic power, with the peak intensity increasing by approximately an order of magnitude. This enhancement, along with lowered threshold RF power required to initiate the 2f peak, indicates the presence of a self-parametric excitation effect. Figure lOe which illustrates the extracted second harmonic (2f) peak power (P2t) for a single SR and series-connected SRs summarizes this trend, showing the lower power threshold and the amplified second harmonic response for the series-connected configuration. The emergence of second harmonic signals can be explained by considering the time-dependent oscillation of magnetoresistance at the same frequency as the input RF current, described by / ?(t) = A / ?ssin (mt + <pB), where cpBis defined as the phase shift. The rectified voltage generated by the SR may be defined as:. . . equation (1) where ARSis defined as the oscillating component of the magnetoresistance, Iacis defined as the amplitude of the RF current driving the SR device, <hsis defined as the phase shift associated with the spin torque, m is the angular frequency of the input RF signal and t is the time variable.

[0067] Based on the above, it can be said that SR devices exhibit a large voltage-controlled magnetic anisotropy (VCMA). When multiple SR devices are connected in series, the voltage component at 2®, generated by one SR, can induce parametric excitation in neighbouring SRs,thereby amplifying the magnetization precession angle. This self-parametric excitation phenomenon significantly enhances the overall rectified voltage (i.e., high sensitivity) as illustrated in Figure 9b. To substantiate this, a systematic micromagnetic simulation was conducted to analyse the dependence of magnetization dynamics on the VCMA amplitude, under conditions where the SR device is driven by an RF current at a frequency ® and simultaneously subjected to a VCMA field oscillating at 2<n (i.e., to simulate an oscillating voltage component generated by an adjacent SR in a series connected array). The results confirm that increasing the VCMA field enhances the amplitude of magnetization precession, thereby supporting the presence of self-parametric excitation and its role in boosting sensitivity. Furthermore, it is anticipated that the microwave magnetic field associated with incident electromagnetic waves may also couple directly with the free magnetic layer, further increasing the precession amplitude and contributing to enhanced rectification.

[0068] In embodiments of the disclosure, an energy harvesting module (EHM) comprising a plurality of 80 x 200 nm2SRs that are connected in series is illustrated in Figure I la. As shown, Figure 1 la illustrates EHM 1102 which comprises a plurality of SRs that are connected in series. The rectified voltage (Vr) generated by the plurality of SRs that are connected in series is initially stored in capacitor 1103 and subsequently stepped up to a higher voltage (Vstep ~ 1.6 - 4 V) using boost converter 1104 to power an electronic device 1105 such as, but not limited to, a temperature sensor.

[0069] While ten SRs are depicted in the configuration, it will be understood by those skilled in the art that any suitable number of SRs may be connected in series within EHM 1102. Due to the high sensitivity of the plurality of 80 x 200 nm2SRs in EHM 1102, ambient RF energy may be harvested using EHM 1 102 without the use of an external antenna.

[0070] Figure 1 lb illustrates the peak rectified voltages (Vr) generated by EHM 1102 when excited using RF signals transmitted by 2.45 GHz and 3.5 GHz antennas. It should be noted that the dashed lines at Vr~ 20 mV in Figure 1 lb represents the threshold voltage for the ON- state of boost converter 1104. Based on the results plotted in Figure 11b, it was determined that EHM 1102 was able to produce a rectified voltage (Vr) of 20 mV at 2.45 GHz and 11 mV at 3.5 GHz when the input RF power, Prf = -25 dBm. Notably, the plurality of SRs connectedin series was able to generate the 20-mV threshold voltage required to activate boost converter 1104 using an input RF power of -22 dBm at 2.45 GHz.

[0071] Once activated, boost converter 1104 boosts the rectified voltage (Vr) generated by EHM 1102 from an initial range of 20-50 mV to 1.6—4 V. The integrated temperature sensor then activates at a stepped-up voltage (Vstep) of approximately 1.2 V, which is achieved at input power levels of -27 dBm using dual RF sources (2.45 GHz and 3.5 GHz), or -22 dBm using a single 2.45 GHz source. The total time required to initially power the sensor is approximately 15-30 seconds. Notably, EHM 1 102 is able to reach the 20 mV rectification threshold within a few seconds, owing to its inherently low capacitance (on the order of picofarads to femtofarads), which is a significant advantage of SR devices designed in accordance with embodiments of the disclosure.

[0072] However, to mitigate the rapid charge-discharge behaviour associated with low intrinsic capacitance, capacitor 1103 having an external 0.01 F capacitor (rated for 3.3 V) is employed to stabilize and store the rectified output. Under dual-source excitation at -25 dBm, EHM module 1102 is able to produce a rectified voltage Vrof ~28 mV, which is stepped up to ~3.7 V. The estimated time to charge the external capacitor to 20 mV under these conditions is approximately 23-24 seconds, closely matching the experimentally observed 20-25 seconds required to activate the sensor. In this embodiment, the differential resistance of the SR-array in EHM 1102 is measured to be approximately 2 k . Once charged, EHM 1102 is able to operate continuously for up to one hour, with the stepped-up voltage Vstep being stored in capacitor 1 103.

[0073] Figure 11c illustrates the discharge behaviour of the stored stepped-up voltage Vstep at an input power level of Prf = -25 dBm (from two RF sources), across different electrical loads. Specifically, the plots in Figure lie show that the stored voltage demonstrates negligible leakage when connected to a low-current handheld multimeter, even after the RF source has been turned off. The plots also show that a temperature sensor (1.2 V) and LED (1.6 V) remain powered for approximately 50 and 30 seconds, respectively, after deactivation of the RF source, thereby demonstrating the reliability, highlighting the ability of EHM 1102 to function reliably in intermittent RF environments. These findings underscore the broadbandoperational flexibility and practical utility of an SR-array that is connected in series for powering a variety of electronic devices.

[0074] It should be noted that the broadband rectification measurements presented in Figures 1 la-11c were conducted using a resonant patch antenna operating in the 2.4-2.5 GHz range (gain ~7.2 dBi) and a whip antenna covering 3.5-3.6 GHz (gain ~3 dBi), both driven by a signal generator. During operation, the rectified voltage output (Vr) from the SR-array is stored in the external capacitor and subsequently amplified by the boost converter to a voltage in the range of 1 .5-4 V. The EHM measurements were conducted with the antenna positioned at variable distances of 2.5-5 cm from the SR-array chip, with a minimum distance of approximately 2.5 cm maintained during testing at an input RF power level of -27 dBm.

[0075] Figure 12a illustrates the measured conversion efficiency, T|(%), of Schottky diodes, SR devices that are connected in series (SR-array), and spin rectifier connected with an impedance matched patch antenna (SR-rectenna) at 2.45 GHz with varying input power levels Prf, while Figure 12b illustrates the sensitivity, (S) of these similar devices at 2.45 GHz with varying input power levels Prf. The Schottky diodes used to generate the plots in Figure 12a comprised of two commercially available low-power Schottky diodes (HSMS-2860 and SMS- 7630) that are each connected to a 50 Ohm antenna and all measurements were made under identical ambient conditions. For clarity, it should be noted that the input RF power (Prf) refers to the output power from the signal generator used in the measurement setup and does not account for antenna gain or impedance mismatch losses at the respective device interfaces. This approach ensures that the actual front-end rectification efficiency is accurately assessed.

[0076] The measured rectification efficiency (q) is defined as 17 =Pdc' / prwhere Pdc= and where Rzbrdenotes the differential resistance of the SR devices or Schottkydiodes, as measured during the measurement step. For a consistent and fair comparison across all evaluated systems, Prf is taken as the actual power output from the RF source (signal generator), without accounting for any transmitting antenna gain, interface losses, or attenuation, thus representing the front-end efficiency. Independent assessments using a vector network analyzer (via S21 parameters), a spectrum analyzer with a receiving antenna, and an RF power meter to capture the radiated power from the transmitting antenna confirmthat the received power at a distance of 2.5 cm deviates from the source power (Prf) by only ±0.5 to 3 dBm This small discrepancy indicates high directivity of the transmitting antenna and negligible loss in the near-field region. Accordingly, for near-field measurements, the received power at 2.5 cm is effectively equivalent to the power emitted by the signal generator, and no correction for antenna gain or interface losses was applied in estimating the sensitivity, irrespective of whether SR-rectennas, SR-arrays, or Schottky diodes are used in the wireless energy harvesting measurements.

[0077] As losses at the device interface are not accounted for in these evaluations, the derived sensitivity values represent the lower bound of the device performance. The measured performances of the HSMS-2860 and SMS-7630 Schottky diodes are consistent with their respective datasheet specifications, validating the accuracy of the measurement protocol. The SR-array measurements were conducted without the use of an external antenna; instead, RF signal coupling is facilitated through a compact ground-signal-ground co-planar waveguide (240 pm x 500 pm) fabricated using Cr (5 nm) / Au (100 nm) metallization via photolithography and lift-off on the SR device. Although short wirebonds (< 1 mm) are used to connect the SR devices, their contribution to RF power coupling is minimal and they primarily serve as a DC or low-frequency transmission path for the rectified output signal. Given that the rectified voltage is sensitive to the zero-bias differential resistance (R / hr) under RF excitation, RZbr was measured for individual SRs, SR-arrays, HSMS-2860, and SMS-7630 diodes under 2.45 GHz irradiation at Prf = -25 dBm. The respective values were found to be approximately 0.27 kQ, 2 kQ, 5 kQ, and 5.1 kQ respectively.

[0078] The experimental results demonstrate that the SR-rectenna operates reliably within the -62 dBm to -20 dBm power range thereby validating its high sensitivity and effectiveness under weak ambient RF conditions. The results plotted in Figures 12a and 12b show that the SR-array outperforms the other devices in the -50 dBm to -20 dBm range by achieving higher rectification efficiency (peaking at p —7.81% at -30 dBm) and exceptional sensitivity (reaching -34,000 mV mW at -50 dBm). These plots also show that when RF energy is harvested at high RF power, i.e., Pii > -20 dBm, the performances of SR devices is limited due to the saturation of rectified voltage at this RF power range.

[0079] While state-of-the-art RF rectifiers based on materials such as Si, GaAs, or M0S2 have demonstrated high efficiencies ranging from 40% to 80% at relatively high input RF powers (Prf > -10 dBm), their effectiveness significantly diminishes at lower RF power levels (Prf < -20 dBm), thereby limiting their applicability in EHM applications. The efficiencies of various state-of-the-art RF rectifiers and SR array are set out in Table 1 below.Table 1

[0080] From Table 1 above, it can be seen that only a limited number of state-of-the-art RF rectifiers were able to achieve efficiencies between 5% and 19% within the - 30 dBm < Prf <- 20 dBm range, after taking into account the efficiency of the antenna connected to the respective RF rectifier Unlike the state-of-the-art RF rectifiers, the SR-array designed in accordance with embodiments of the present disclosure was able to achieve comparable efficiency measurements even when antenna contributions were not included.

[0081] It should be noted that the SR-array may be integrated within a compact equivalent electrical footprint of approximately 1 mm2, inclusive of the co-planar waveguides used to couple RF power to the respective SR devices that are connected in series. In contrast, commercially packaged Schottky diodes typically occupy over 10 mm2, and when combined with a high-efficiency antenna, the total area extends to several square centimetres, rendering them unsuitable for on-chip integration. Even with the use of miniaturized antennas, theoverall footprint remains around 200 mm2for low-power applications. In light of this, the SR- array offers a highly sensitive and space-efficient solution for ambient energy harvesting, and its potential for further miniaturization and enhanced efficiency is expected to increase with advances in on-chip antenna technologies.

[0082] Based on the plots in Figure 12b, it can be seen that the SR-rectenna exhibits an exceptional RF sensitivity of approximately 10,000 mV mW1at -62 dBm, enabling reliable RF energy harvesting in weak and noisy ambient environments. For single SR devices, intrinsic material and structural properties, such as perpendicular magnetic anisotropy, device geometry, and the dipolar field from the polarizer layer, define the energy landscape of the nanomagnet, thereby facilitating large-angle magnetization precession at ultralow RF input power. This enhanced sensitivity is governed by the dynamic behaviour of the magnetic tunnel junction (MTJ) and is influenced by both the zero-field TMR and the VCMA coefficient, which together boost the zero-bias rectified voltage.

[0083] Additionally, external impedance matching structures contribute to further sensitivity gains in SR-rectennas by reducing power loss. In the SR-array configuration, a VCMA-driven self-parametric excitation mechanism enhances both sensitivity and detection bandwidth without requiring an external antenna or matching network. Leveraging these features, an on-chip co-planar waveguide-based SR-array energy harvesting module may be used to power commercial temperature sensors at RF power levels as low as -27 dBm. In summary, EHM based on SR-devices as described in this disclosure are a superior alternative to conventional rectifiers in ultralow-power applications, offering a combination of high efficiency, minimal parasitic effects, ease of integration, and scalability.

[0084] Micromagnetic Simulations

[0085] Micromagnetic simulations were performed by numerically solving the Landau-Lifshitz-Gilbert-Slonczewski equation:where aGis defined as the Gilbert damping, m — M / Msis defined as the normalized magnetization vector, y is the gyromagnetic ratio and Msis the saturation magnetization of theMTJ free layer (FL). The effective magnetic field, he^, includes the demagnetizing field, and the interfacial uniaxial perpendicular anisotropy. The spin-transfer torque (STT) term is proportional to the pre-factor a where g is the gyromagnetic splitting factor, gBis the Bohr magneton, e is the electron charge, dzis the thickness of the FL, and P is the spin polarization. The alternating current density flowing into the device is given by J = / QCsin (2nft + (Pi) The VCMA is included in the effective field as a changing anisotropy field.

[0086] In embodiments of the disclosure, an elliptical 40 nm x 100 nm x 1 nm hybrid device was simulated, where it was then determined that the perpendicular anisotropy was larger than the demagnetizing magnetic field and hence the equilibrium direction of the magnetization m is tilted from the out-of-plane direction as observed in the experiment. The polarizer magnetization p is in-plane along the -x direction. The main micromagnetic parameters are Ms= 800 \iA / m, perpendicular uniaxial anisotropy constant Ku= 0.39 MJ / m3, aG— 0.02, g = 2 , and P — 0.7 , and the dipolar field originated by the polarizer being added as a constant HDC= -5 mT. In simulations, it was found that the rectification curves are computed for the amplitude of the alternating current density raging from Jac— 0.1 to 10 MA / cm2and a frequency scan from f — 0.1 to 15 GHz.

[0087] Optimization of Interfacial Anisotropy

[0088] The optimization of MgO (tunnel barrier) / CoFeB (free layer) interface in magnetic tunnel junction (MTJ) based spin-rectifiers (SRs) are required to manipulate the interfacial perpendicular anisotropy (IP A) and to achieve a high rectification sensitivity without external magnetic field and DC bias. For this purpose, the CoFeB (tcoFen) and MgO (tMgo) thicknesses are varied to tune the IPA.

[0089] Figure 13a illustrates the tunnelling magnetoresistance (TMR) of the spin rectifier device as a function of the spacer layer thickness, for free magnetic layers having a thickness of 1.7 nm, 1.8 nm and 1.9 nm, respectively. Figure 13b illustrates the resonance frequencycorresponding to the maximum rectification value of the spin rectifier device as a function of the spacer layer thickness, for free magnetic layers having a thickness of 1.7 nm, 1.8 nm and 1.9 nm, respectively. Figure 13c illustrates the RF sensitivity at the resonance frequency of the spin rectifier device as a function of the spacer layer thickness, for free magnetic layers having a thickness of 1.7 nm, 1.8 nm and 1.9 nm, respectively.

[0090] From the plots in Figures 13a-13c, it is observed that a canted equilibrium direction due to moderate interfacial perpendicular anisotropy (IP A) occurs when a thickness of the free magnetic layer tcoFeB = 1 .9 nm and a thickness of the spacer layer twgo = 1 nm. Based on these thicknesses, the SR device was able to achieve a high tunnelling magnetoresistance (as shown in Figure 13a), a resonance frequency close to ~2 GHz (as shown in Figure 13b) and a high sensitivity (as shown in Figure 13c) when an elliptical dimension of 80 x 200 nm2was adopted

[0091] Optimization of the canted equilibrium direction of SRs

[0092] In addition to optimizing the thicknesses of the spacer layer and the free magnetic layer, the junction dimensions of the SRs were also optimized to further control the equilibrium direction of the out-of-plane tilted magnetization at zero field. The canted equilibrium angle primarily arises from the competition of in-plane shape anisotropy and out- of-plane interfacial anisotropy. The canted equilibrium angle ( 6tiit) between the magnetization of the in-plane reference magnetic layer and that of the tilted free magnetic layer can be estimated by:...equation (3) R(9tut), Rp,ar|d RAP are the zerofield, parallel, and anti-parallel configuration resistances, respectively. Figure 14a illustrates the magnetoresistance of a 40 * 100 nm2spin rectifier, demonstrating a canted equilibrium magnetization resulting from the competition between inplane shape anisotropy and out-of-plane interfacial anisotropy and Figure 14b illustrates the magnetoresistance of a 80 x 200 nm2spin rectifier, demonstrating a canted equilibrium magnetization resulting from the competition between in-plane shape anisotropy and out-of- plane interfacial anisotropy.

[0093] A significant canted equilibrium angle of 48° and 80° was obtained for the smalldimension devices: 40 x 100 nm2to 80 x 200 nm2, respectively. For bigger dimension devices (>100 x 250 nm2), the out-of-plane anisotropy is significantly low, favouring in-plane easy- axis anisotropy. Also, the zero-field TMR (AR?b = (Rzb - RP) / RP) reaches 30-40% in some of the 80 x 200 nm2SR devices as shown in Figure 14b, resulting into a high rectified voltage. Here, RZb is defined as the zero-field resistance.

[0094] Optimization of SR size for the best RF sensitivity

[0095] Figure 15a illustrates the tunnelling magnetoresistance (TMR) and the resistancearea (RA) product as a function of the junction dimension of the spin rectifier device, Figure 15b illustrates canted equilibrium angle as a function of the junction dimension of the spin rectifier device, Figure 15c illustrates the zero-bias rectification signal at an input power signal of Prf = -30 dBm as a function of input frequency, for junction dimensions of 120 x 300 nm2(xlO), 100 x 250 nm2, 80 x 200 nm2, 60 x 150 nm2, and 40 x 100 nm2, and Figure 15d illustrates the frequency of peak rectification voltage and corresponding zero bias RF- sensitivity as a function of the junction dimension of the spin rectifier device.

[0096] The SRs with dimensions 80 x 200 nm2using a free magnetic layer having a thickness tcoteB = 1.9 nm and a spacer layer having a thickness tMgo = 1 nm showed the best TMR of -75% and a resistance-area (RA) product of -5 Q pm2(as shown in Figure 15a). Other characteristics are a canted equilibrium angle of 79-95° (as shown in Figure 15b), a ferromagnetic resonance (FMR) frequency close to 2.4 GHz and a maximum sensitivity of — 1000 mV mW'1at zero bias and zero magnetic field (as shown in Figures 15c and 15d).

[0097] Contribution of the voltage controlled magnetic anisotropy

[0098] The voltage controlled magnetic anisotropy (VCMA) is known to enhance the RF sensitivity in MTJs when a bias (or electric field) is applied across the device, as it excites the voltage induced ferromagnetic resonance leading to an extra antisymmetric term in the rectification spectra. The amplitude of VCMA can be calculated as the variation of theeffective magnetic anisotropy energy per unit volume of the free layer (Ep) from the area under the hysteresis loopM(H). Specifically,...equation (4) where Hp is the effective perpendicular anisotropy field. Figure 16a illustrates the shift of the perpendicular anisotropic field within the spin rectifier device as a function of the DC bias voltage, for junction dimensions of 40x100 nm2, 60 x 150 nm2, 80 x 200 nm2, and 100 x 250 nm2as calculated by the shift in the conductance vs. magnetic field hysteresis loops with DC voltages. Figure 16b illustrates the voltage-controlled magnetic anisotropy (VCMA) coefficient as extracted from the slope of the plots in Figure 16a based on equation (4) above.

[0099] Dependency of rectified voltage on RF input power

[0100] Figure 17 illustrates the rectified voltage as a function of varying RF power, for SR devices having junction dimensions of 40 x 100 nm2and 80 x 200 nm2. From the plots in Figure 17, it can be seen that the rectified voltage, Vr, of the 80 x 200 nm2SR device saturates at a high RF input power, Prf because of the large amplitude magnetization dynamics achieved by this SR device. In this structural configuration, additional power is transferred to high order modes, and this causes the sensitivity to decrease (see Figure 7). The saturation voltage of the 40 x 100 nm2SR device is significantly less because of the device’s lower sensitivity as compared to the 80 x 200 nm2SR device and transition of resonant to broadband rectification response. As discussed in the earlier sections, the sensitivity of these two SR devices decreases from Prf > -20 dBm (see Figure 7) due to the decrease of ARZb and saturation of rectified voltage (Vr) (see Figure 17).

[0101] Matched antenna design and coupling of R and Tx antenna

[0102] Figure 18a illustrates the on-chip impedance of a spin rectifier device based on the reflection coefficient Sn as measured by vector network analyser (VNA) where plot 1801 represents the imaginary part of the impedance while plot 1802 represents the real part of the impedance The drastic change in the impedance around 2.45 GHz is attributed to the involvement of connectors attached to the SR chip. However, the impedance evaluated beforeand after designing the matched antenna shows a robust value. In embodiments of the disclosure, the impedance matching could be improved by designing an on-chip impedance matched antenna.

[0103] Figure 18b illustrates the reflection coefficient of a spin rectifier antenna. From the plot in this Figure, it can be seen that the antenna can cover the entire 2.4-2.5 GHz band with a reflection coefficient (Sn) that is less than -10 dB.

[0104] Figure 18c illustrates the 3D radiation pattern of a spin rectifier antenna designed using a computer simulation. The antenna is coupled with a high-gain of 7.2 dBi in the near- field region as shown by the radiation 3D pattern showing the gain in vertical direction.

[0105] Figure 18d illustrates the received power, as measured by the vector network analyzer (VNA), as a function of the frequency of the incident RF signal when the RF signal is applied from a near-field distance of 2.5 cm, for various input RF power levels. This plot shows the coupling between the receiving and transmitting antenna as estimated by the VNA measurements.

[0106] Figure 18e illustrates the received power, as measured by the VNA, as a function of input RF power (Pre) for various near-field distances. These plots show that the received RF power at 2.45 GHz is very close to the power provided by the signal generator (Prf) in the near field region of 2.5cm. This shows a high directivity of the transmission and receiving antenna.

[0107] Performance of a single SR device for a bandwidth between 0 and 6 GHz.

[0108] Figure 19a illustrates the measured voltage rectification of a spin rectifier device as a function of the frequency under two magnetic field conditions. Plot 1901 illustrates the rectified voltage when a magnetic field of 980 Oe is applied while plot 1902 illustrates the rectified voltage when a magnetic field of 0 Oe is applied. These plots show that the SR device’s RF bandwidth may be tuned by an applied magnetic field. By applying 980 Oe, a large bandwidth of 0.1—6 GHz was achieved using a single 40 x 100 nm2SR device, i.e., plot 1901.

[0109] Figure 1 b illustrates the variation in the peak rectified voltage of the spin rectifier device as a function of the applied magnetic field. The plot in this Figure shows the peak resonant voltage with respect to the external magnetic field applied in-plane along the major axis of the elliptical SRs, which shows that the device works in a wide range of magnetic field. However, the rectification voltage response starts to degrade at magnetic fields > 1200 Oe, because of a low TMR as the free layer becomes magnetically harder at higher magnetic fields.

[0110] Scaling of output voltage as a number of SRs connected in series increases.

[0111] Figure 20a illustrates the peak rectified voltage as a function of the number of spin rectifier devices connected in series for junction dimensions of 40 x 100 nm2and 80 200 nm2. As the plots show, when the number of spin rectifiers increases, the peak response becomes saturated.

[0112] Figure 20b illustrates the rectification bandwidth at which the peak rectification voltage reduces by half as a function of the number of spin rectifier devices connected in series for junction dimensions of 40 x 100 nm2and 80 x 200 nm2. The plots in this Figure show that the rectification bandwidth (BW) increases as the number of SRs increases..

[0113] Effect of VCMA on the amplitude of magnetization precession.

[0114] Figure 21 illustrates the simulated precession amplitude of the x-component of magnetization (Amx) as a function of alternating current density (Jac). The presence of a VCMA field of 10 mT leads to a significantly larger precession amplitude at lower current densities.

[0115] Stability of SR arrays for long operation times.[001 16] Figure 22a illustrates the rectification voltage of a spin rectifier device as a function of time, for RF input signals having various operating frequencies and Figure 22b illustrates the energy harvesting module voltage of a spin rectifier device as a function of time, for RF input signals having various operating frequencies.

[0117] To obtain these results, the voltage from the SR device and EHM module were stored in the nanovoltmeter for one hour. When the source of wireless signal is available consistently in the ambient condition where the RF power is at Prf > -20 dBm, the SR device as well as EHM show a very consistent voltage, and this is shown in plot Figure 22a. This result confirms the absence of the charging-discharging instability, desynchronization among the series-coupled SR devices in series and microwave induced heating effects, all of which that could otherwise affect the SR devices performance over time.[001 18] Signal-to-noise-ratio and noise equivalent power.

[0119] Figure 23a illustrates the signal -to-noise ratio of various spin rectifier designs as a function of the input RF power. The signal-to-noise ratio (SNR) determines the performance of the rf rectifiers in the noisy environment. The SNR is calculated using the formula, SNR = Psigiial / Pnoise, where theare defined as the DC voltages recorded by the nanovoltmeter when the 2.45 GHz wireless signal is ON and OFF, respectively. Rdc is defined as the DC resistance of the devices. Moreover, the noise equivalent power (NEP) of a single SR is estimated, which characterises the noise level in quadratic detectors. The NEP is defined as the ratio of noise voltage and sensitivity. For the measurement of noise voltage (Vnoise) in the spectrum analyser, the measurement setup of measuring a diode mixing noise was adopted. The SR devices having a sensitivity > 1000 mV / mW was adopted for these measurements. The SR-array shows a low SNR at an input RF power Prf < -50 dBm in comparison to the SR-rectenna due to an increased parasitic capacitance. Further, the results show that SR rectenna works reliably (SNR > 20 dB) for -62 dBm < Prf < -20 dBm, enabling SR-rectenna sensitive for the weak ambient conditions.

[0120] Figure 23b illustrates the noise-equivalent-power (NEP) of a spin rectifier device as a function of the input RF power From this plot, the NEP of ~2 x 10'12W Hz°5when the input RF power, P,r 0.1 pW (-40 dBm) and without a DC bias, i.e., Ide = 0 mA, is very close to the NEP of ~8 x 10'12W Hz0 5or 3.6 x 10'12W Hz° " as measured in existing SR devices that require a DC bias. In general, a low NEP is observed in the SR devices because of the high sensitivity, which leads to a good SNR even in SR-array configurations.

[0121] Comparison of the SR device described in this disclosure with rectifiers known the art.

[0122] Figure 24 illustrates the efficiency of various spin rectifier designs as a function of the input RF power. The plots demonstrate that the efficiency achieved by the disclosed SR device is two to three orders of magnitude higher than that reported for prior designs, including both resonant and broadband spin rectifiers.

[0123] Measurement setup of the SR device and EHM.

[0124] Figure 25a illustrates a schematic layout of a co-planer waveguide ground-signal- ground design for use with a spin rectifier device and Figure 25b illustrates a photograph of a co-planer waveguide ground-signal-ground design as implemented on a printed circuit board. The SR device to be measured is positioned in the centre of the schematic of CPW shown in these Figures.

[0125] A process for forming a nanoscale spin rectifier device for harvesting ambient radiofrequency energy is illustrated in Figure 26. Process 2600 begins at step 2602 with process 2600 forming a reference magnetic layer that has a fixed magnetization direction. At step 2604, process 2600 then forms a spacer layer on the reference magnetic layer. Process 2600 then forms a free magnetic layer on the spacer layer at step 2606. A rectified DC voltage is then generated at step 2608 when RF signals are incident on the device. In embodiments of the disclosure, the spacer layer is configured to enable spin dependent tunnelling of electrons between the reference layer, and the free magnetic layer has an equilibrium magnetization direction that is canted relative to the fixed magnetization direction of the reference magnetic layer, the equilibrium magnetization direction being determined by a thickness of the free magnetic layer and a thickness of the spacer layer. It is noted that the RF signals incident on the device induces precessional motion of the magnetization direction of the free magnetic layer which modulates a tunnelling resistance across the spacer layer to generate a rectified DC voltage across the device.

[0126] Process 2600 further comprises the steps of forming a synthetic antiferromagnetic structure below the reference magnetic layer and forming an exchange bias layer below thesynthetic antiferromagnetic structure, wherein the exchange bias layer comprises an antiferromagnetic material, and wherein the reference magnetic layer and the free magnetic layer each comprise a cobalt-iron-boron (CoFeB) alloy, and the synthetic antiferromagnetic structure comprises a multilayer magnetic stack. In embodiments of the disclosure, the multilayer magnetic stack of the synthetic antiferromagnetic structure may comprise a cobalt (Co) layer, a ruthenium (Ru) layer, and a CoFeB layer, wherein the synthetic antiferromagnetic structure is exchange-biased by an antiferromagnetic material layer comprising platinum -manganese (PtMn), iron-manganese (FeMn), or iridium-manganese (IrMn).

[0127] In embodiments of the disclosure, a process may be provided for providing a spin rectifier rectenna system for harvesting ambient radiofrequency energy. The process comprises the step of electrically coupling a high-gain antenna to the SR device formed as described in the previous sections above, wherein the high-gain antenna is configured to be impedance matched to an operating frequency of the SR device.

[0128] In embodiments of the disclosure, a process may be provided for providing an ambient radiofrequency energy harvesting module. The process comprises the steps of connecting a plurality of the SR devices formed as described in the previous sections above in a series, parallel or hybrid-series-parallel configuration, and coupling a pair of output terminals of the module to an electrical load.

[0129] In embodiments of the disclosure for the spin rectifier devices connected in the series configuration, a total rectified DC voltage across the electrical load increases with a number of the plurality of devices connected in series, while for spin rectifier devices connected in the parallel configuration, and a total rectified direct current (DC) across the electrical load increases with a number of the plurality of devices connected in parallel.

[0130] Numerous other changes, substitutions, variations, and modifications may be ascertained by the skilled in the art and it is intended that the present application encompass all such changes, substitutions, variations, and modifications as falling within the scope of the appended claims.

Claims

CLAIMS1. Ananoscale spin rectifier device for harvesting ambient radiofrequency energy comprising: a reference magnetic layer having a fixed magnetization direction; a spacer layer disposed on the reference magnetic layer, the spacer layer configured to enable spin dependent tunnelling of electrons between the reference layer and a free magnetic layer disposed on the spacer layer, wherein the free magnetic layer has an equilibrium magnetization direction that is canted relative to the fixed magnetization direction of the reference magnetic layer, the equilibrium magnetization direction being determined by a thickness of the free magnetic layer and a thickness of the spacer layer, and wherein RF signals incident on the device induces precessional motion of the magnetization direction of the free magnetic layer which modulates a tunnelling resistance across the spacer layer to generate a rectified DC voltage across the device.

2. The device according to claim 1, wherein the generation of the rectified DC voltage occurs without requiring an external direct-current (DC) bias or magnetic field to be applied across the device.

3. The device according to claim 1 further comprising: a synthetic antiferromagnetic structure disposed below the reference magnetic layer and an exchange bias layer being disposed below the synthetic antiferromagnetic structure, the exchange bias layer formed from an antiferromagnetic material; wherein the reference magnetic layer and the free magnetic layer each comprise a cobalt-iron-boron (CoFeB) alloy, and the synthetic antiferromagnetic structure comprises a multilayer magnetic stack.

4. The device according to claim 3, wherein the multilayer magnetic stack of the synthetic antiferromagnetic structure comprises a cobalt (Co) layer, a ruthenium (Ru) layer, and a CoFeB layer, wherein the synthetic antiferromagnetic structure is exchange-biased by an antiferromagnetic material layer comprising platinum-manganese (PtMn), iron-manganese (FeMn), or iridium-manganese (IrMn).

5. The device according to any one of claims 1 to 4, wherein the spacer layer comprises magnesium oxide (MgO).

6. The device according to any one of claims 1 to 5, wherein the free magnetic layer comprises a thickness between 1 nm and 6 nm, the spacer layer comprises a thickness between 0.6 nm and 3 nm, and the device comprises lateral dimensions between 40 X 50 nm2and 80 X 200 nm2.

7. The device according to claim 6, wherein the device comprises a horizontal elliptical cross- sectional shape that has a major axis and a minor axis with an aspect ratio of 1 to 5.

8. A spin rectifier rectenna system for harvesting ambient radiofrequency energy, the system comprising: a high-gain antenna that is electrically coupled to the device according to any one of claims 1 to 7, wherein the high-gain antenna is configured to be impedance matched to an operating frequency of the device.

9. An ambient radiofrequency energy harvesting module comprising: a plurality of the devices according to any one of claims 1 to 7, the plurality of devices being connected in a series configuration and having a pair of output terminals coupled to an electrical load.

10. An ambient radiofrequency energy harvesting module comprising: a plurality of the devices according to any one of claims 1 to 7, the plurality of devices being connected in a parallel configuration and having a pair of output terminals coupled to an electrical load1 1 . An ambient radiofrequency energy harvesting module comprising: a plurality of the devices according to any one of claims 1 to 7, the plurality of devices being connected in a hybrid series and parallel configuration and having a pair of output terminals coupled to an electrical load.

12. The module according to any one of claims 9 to 11, wherein the plurality of the devices comprises at least ten nanoscale spin rectifier devices.

13. The module according to any one of claims 9 to 11, wherein each of the plurality of devices is electromagnetically coupled to an incident RF signal by a respective co-planar waveguide, each co-planar waveguide configured to receive the RF signals and to provide the received RF signals to a corresponding device.

14. The module according to claim 9, wherein the plurality of the devices comprises between two and ten nanoscale spin rectifier devices connected in the series configuration, and wherein a total rectified DC voltage across the electrical load increases with a number of the plurality of devices connected in series.

15. The module according to claim 10, wherein the plurality of the devices comprises between two and ten nanoscale spin rectifier devices connected in the parallel configuration, and wherein a total rectified direct current (DC) across the electrical load increases with a number of the plurality of devices connected in parallel.

16. A method for forming a nanoscale spin rectifier device for harvesting ambient radiofrequency energy, the method comprising: forming a reference magnetic layer having a fixed magnetization direction; forming a spacer layer on the reference magnetic layer; and forming a free magnetic layer on the spacer layer, wherein the spacer layer is configured to enable spin dependent tunnelling of electrons between the reference layer, wherein the free magnetic layer has an equilibrium magnetization direction that is canted relative to the fixed magnetization direction of the reference magnetic layer, the equilibrium magnetization direction being determined by a thickness of the free magnetic layer and a thickness of the spacer layer, and wherein RF signals incident on the device induces precessional motion of the magnetization direction of the free magnetic layer which modulates a tunnelling resistance across the spacer layer to generate a rectified DC voltage across the device.

17. The method according to claim 16, wherein the generation of the rectified DC voltage occurs without requiring an external direct-current (DC) bias or magnetic field to be applied across the device.

18. The method according to claim 16 further comprising: forming a synthetic antiferromagnetic structure below the reference magnetic layer; and forming an exchange bias layer below the synthetic antiferromagnetic structure, wherein the exchange bias layer comprises an antiferromagnetic material, wherein the reference magnetic layer and the free magnetic layer each comprise a cobalt-iron-boron (CoFeB) alloy, and the synthetic antiferromagnetic structure comprises a multilayer magnetic stack.

19. The method according to claim 18, wherein the multilayer magnetic stack of the synthetic antiferromagnetic structure comprises a cobalt (Co) layer, a ruthenium (Ru) layer, and a CoFeB layer, wherein the synthetic antiferromagnetic structure is exchange-biased by an antiferromagnetic material layer comprising platinum-manganese (PtMn), iron-manganese (FeMn), or iridium-manganese (IrMn).

20. The method according to any one of claims 16 to 19, wherein the spacer layer comprises magnesium oxide (MgO).

21. The method according to any one of claims 16 to 20, wherein the free magnetic layer comprises a thickness between 1 nm and 6 nm, the spacer layer comprises a thickness between 0.6 nm and 3 nm, and the device comprises lateral dimensions between 40 x 50 nm2and 80 x 200 nm2.

22. The method according to claim 21, wherein the device is formed to have a horizontal elliptical cross-sectional shape that has a major axis and a minor axis with an aspect ratio of 1 to 5.

23. A method for providing a spin rectifier rectenna system for harvesting ambient radiofrequency energy, the method comprising: electrically coupling a high-gain antenna to the device formed according to any one of claims 16 to 22, wherein the high-gain antenna is configured to be impedance matched to an operating frequency of the device.

24. A method for providing an ambient radiofrequency energy harvesting module, the method comprising: connecting a plurality of the devices according to any one of claims 16 to 22 in a series configuration; and coupling a pair of output terminals of the module to an electrical load.

25. A method for providing an ambient radiofrequency energy harvesting module, the method comprising: connecting a plurality of the devices according to any one of claims 16 to 22 in a parallel configuration; and coupling a pair of output terminals of the module to an electrical load.

26. A method for providing an ambient radiofrequency energy harvesting module, the method comprising: connecting a plurality of the devices according to any one of claims 16 to 22 in a hybrid series and parallel configuration; and coupling a pair of output terminals of the module to an electrical load.

27. The method according to any one of claims 24 to 26, wherein the plurality of the devices comprises at least ten nanoscale spin rectifier devices.

28. The method according to any one of claims 24 to 26, further comprising the step of: electromagnetically coupling each of the plurality of devices to a respective co-planar waveguide configured to receive the RF signals and to provide the received RF signals to a corresponding device.

29. The method according to claim 24, wherein the plurality of the devices comprises between two and ten nanoscale spin rectifier devices connected in the series configuration, and wherein a total rectified DC voltage across the electrical load increases with a number of the plurality of devices connected in series.

30. The method according to claim 25, wherein the plurality of the devices comprises between two and ten nanoscale spin rectifier devices connected in the parallel configuration, andwherein a total rectified direct current (DC) across the electrical load increases with a number of the plurality of devices connected in parallel.

Images