Method for regulating and controlling perpendicular magnetic anisotropy and magnetic ion effect of magnetic film through neon ion implantation

Abstract

The invention discloses a method for regulating and controlling the perpendicular magnetic anisotropy and the magnetic ion effect of a magnetic film through neon ion implantation. Belongs to the technical field of magnetic material preparation and optimization. The inert neon ions are selected to carry out irradiation pretreatment on the magnetic thin film, so that the vertical magnetic anisotropy of the thin film is remarkably enhanced, and a solid foundation is laid for manufacturing a high-stability and high-density storage unit. Besides, the film pretreated by neon ions successfully induces a multi-stage and reversible magnetic ion effect: oxygen ions (O < 2->) can be driven to controllably migrate in the film by applying a specific grid voltage, so that efficient and reversible switching of magnetic anisotropy between an in-plane (IMA) state and a vertical (PMA) state is realized, the response speed is high, and the magnetic anisotropy is high in performance. And a brand new implementation way is provided for the design of low-power-consumption nonvolatile magnetic memories and logic devices.

Description

Technical Field

[0001] This invention relates to the field of magnetic material preparation and optimization technology, and more specifically to a method for controlling the vertical magnetic anisotropy and magnetic ion effect of magnetic thin films by neon ion implantation. Background Technology

[0002] With the advent of the era of artificial intelligence and big data, traditional computing systems based on the von Neumann architecture face severe challenges such as the "memory wall" bottleneck, high power consumption, and low efficiency due to their inherent separation of memory and computation. Spintronic devices, especially magnetic random access memory (MRAM), are considered the core hardware foundation for building the next-generation computing paradigm due to their non-volatility, high speed, high durability, and the potential for in-memory computing. The core unit of MRAM is the magnetic tunnel junction (MTJ), which consists of a thin tunneling barrier layer sandwiched between two ferromagnetic (FM) layers. Its resistance state depends on the relative orientation (parallel or antiparallel) of the magnetization directions of the two ferromagnetic layers. Traditional data writing methods rely on the spin-transfer torque (STT) or spin-orbit torque (SOT) effect, using a large current to drive magnetization switching, but this leads to high operating power consumption and thermal management problems.

[0003] To overcome the aforementioned limitations, voltage-controlled magnetism (VCM) technology has emerged. This technology uses an electric field, rather than a current, to modulate magnetic properties, potentially achieving orders-of-magnitude reductions in energy consumption. While the VCM effect based on interfacial charge accumulation / depletion offers a fast response, it is volatile. Magneto-ionics, as an emerging voltage-controlled magnetism mechanism, uses voltage to drive ions (such as H+). + Li + N 3- O 2- F - (etc.) migrate and undergo electrochemical reactions within the material, thereby achieving non-volatile and highly reversible control over properties such as magnetic anisotropy and saturation magnetization, opening up new avenues for low-power, non-volatile magnetic storage and logic devices.

[0004] Despite the enormous potential of the magneto-ion effect, its practical applications still face numerous challenges, as its performance is strongly dependent on the microstructure of the material system. In typical heavy metal (HM) / ferromagnetic (FM) / metal oxide (MO) multilayer film structures (such as Ta / CoFeB / MgO), the interface atomic configuration, crystal quality, and defects (such as grain boundaries and vacancies) not only determine the basic magnetic properties but also profoundly influence the migration barriers and transport channels of ions, thus dominating the efficiency, speed, and reversibility of the magneto-ion effect.

[0005] Currently, most research in this field focuses on exploring different reactive ions (such as O). 2-H + The migration behavior of magnetic ions and their modulation of magnetism are studied. However, research on how to actively and effectively modify the material matrix itself to enhance its magnetic ion response is still insufficient. Ion irradiation / implantation technology, as a mature material modification method, has been proven to effectively control the anisotropy and Dzyaloshinskii-Moriya interaction (DMI) properties of magnetic thin films. However, existing technologies have the following limitations:

[0006] First, existing studies mostly use active ions (such as N) + O + Implantation introduces physical damage and inevitably introduces complex chemical changes (such as the formation of nitrides and oxides). These chemical effects are coupled with the desired physical modification effects, making the process window difficult to control and potentially having an adverse impact on the long-term stability of the device.

[0007] Secondly, in complex magnetic ion systems composed of ultrathin layers (typically on the nanometer scale), how to precisely and selectively optimize the state of key interfaces (such as CoFeB / MgO) through ion implantation while avoiding damage to other layers or interfaces remains a challenging problem that has not yet been systematically solved. Simple implantation may lead to excessive mixing and severe interlayer diffusion at the interface, thereby destroying the interfacial properties that are crucial to the magnetic ion effect.

[0008] Third, and more importantly, existing technologies lack a universal, quantifiable physical parameter to correlate ion implantation processes with the final magnetic properties and magneto-ion effect performance. This makes process optimization largely dependent on trial and error, making predictive design and controllable fabrication difficult.

[0009] Therefore, there is an urgent need in the field for a technical solution that can overcome the above-mentioned shortcomings, namely, to provide an ion implantation method and a method for controlling the magnetic ion effect that can modify materials purely through physical processes and achieve a systematic enhancement of the magnetic ion effect. This is a technical problem that needs to be solved by those skilled in the art. Summary of the Invention

[0010] In view of this, the present invention provides a method for controlling the vertical magnetic anisotropy and magnetic ion effect of magnetic thin films by neon ion implantation. Through physical modification, the vertical magnetic anisotropy of the magnetic thin film is successfully enhanced, and the reversible control of the magnetic ion effect is successfully achieved by voltage driving.

[0011] To achieve the above objectives, this application adopts the following technical solution:

[0012] The primary objective of this application is to provide a method for controlling the perpendicular magnetic anisotropy of a magnetic thin film using neon ion implantation, comprising the following steps:

[0013] (1) Deposit a Ta / CoFeB / Pt / MgO / HfO2 multilayer film on the substrate;

[0014] (2) Neon ions were used at an energy of 5-25 keV and a flux of 2.5 × 10⁻⁶. 13 –2.0×10 14 Irradiation implantation of the multilayer film described in step (1) under the condition of ions / cm² induces soft atom mixing, Pt layer dissolution and CoFeB layer crystal structure change at the CoFeB / Pt / MgO interface and CoFeB / Ta interface, thereby achieving the enhancement of vertical magnetic anisotropy.

[0015] As a preferred technical solution, the angle of the neon ion implantation is 2–90° relative to the normal of the sample surface.

[0016] As a preferred technical solution, the neon ion flux is 7.5×10¹³ ions / cm², the energy is 15 keV, and the injection angle is 7° relative to the sample surface normal.

[0017] Another object of this application is to provide a method for inducing and modulating magnetic ion effects in magnetic thin films, comprising the following steps:

[0018] (1) The Ta / CoFeB / Pt / MgO / HfO2 multilayer film was pretreated with neon ion implantation as described above;

[0019] (2) Apply voltage to the injected sample to achieve multi-stage reversible control of magnetic anisotropy by driving oxygen ion migration.

[0020] As a preferred technical solution, the voltage regulation process is as follows:

[0021] (1) Apply a negative voltage from -1.0 V to -3.5 V to achieve the transformation from in-plane magnetic anisotropy to maximum perpendicular magnetic anisotropy;

[0022] (2) Based on step (1), continue to apply a negative voltage to -4.0 V to realize the transformation from perpendicular magnetic anisotropy to in-plane magnetic anisotropy;

[0023] (3) Based on step (2), apply a positive voltage from +1.0 V to +4.5 V to restore the perpendicular magnetic anisotropy.

[0024] As a preferred technical solution, the reversible conversion between vertical magnetic anisotropy and in-plane magnetic anisotropy in steps (2) and (3) is achieved through the contamination and decontamination mechanism of the CoFeB / Ta interface.

[0025] Another object of this application is to provide a magnetic structure comprising a Ta / CoFeB / Pt / MgO / HfO2 multilayer film treated by the method described above, the multilayer film exhibiting enhanced perpendicular magnetic anisotropy and / or voltage-tunable magnetic ionization.

[0026] Another object of this application is to provide: the application of the magnetic structure in magnetic storage devices or magnetic logic devices.

[0027] Another object of this application is to provide a magnetic random access memory comprising the magnetic structure as a free layer.

[0028] As can be seen from the above technical solution, compared with the prior art, the present invention has the following beneficial effects:

[0029] (1) This application uses inert neon ions for irradiation, which avoids the complex chemical reactions that may be introduced by active ions, ensures the pure physical properties of the control process, and improves the purity and reliability of the process. The pretreatment of Ta / CoFeB / Pt / MgO / HfO2 multilayer films by the above method in this application significantly enhances the perpendicular magnetic anisotropy (PMA), laying a solid foundation for the fabrication of high-stability, high-density memory cells.

[0030] (2) Furthermore, the thin film pretreated with neon ions successfully induced a multi-stage, reversible magneto-ionic effect: by applying a specific gate voltage, oxygen ions (O2O3) could be driven to ionize. 2- Controllable migration in thin films enables efficient and reversible switching of magnetic anisotropy between in-plane (IMA) and perpendicular (PMA) states with fast response speed, providing a new approach for the design of low-power, non-volatile magnetic memories and logic devices.

[0031] (3) The present invention also has high process controllability and predictability: by introducing the per-atom displacement (DPA) value to quantify irradiation damage, a quantitative correlation between irradiation parameters (energy, flux) and final magnetic properties is established, which makes process optimization shift from the traditional "trial and error" mode to a predictable and designable model-driven approach.

[0032] (4) The entire technical solution is highly compatible with existing semiconductor processes. Ion implantation is a mature standard process in the semiconductor industry, so this technology is easy to integrate into existing microelectronic manufacturing processes, especially for improving the performance and industrial application of spintronic devices such as magnetic random access memory (MRAM).

[0033] In summary, this application achieves significant performance improvements, precise process control, and good industrial compatibility through physical modification via inert ion implantation combined with reversible magnetic ion effect manipulation, providing strong technical support for the development of next-generation spintronic devices. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0035] Figure 1 For: the comparison results of unirradiated samples and samples irradiated with neon ions at different fluxes; where a)-k) is 15 keV Ne + The irradiated sample was at 2.5 × 10⁻⁶. 13 - 2.5×10 14 ions / cm² Hysteresis loops at various flux levels; l) Relationship between saturation magnetization and ion flux; m) Relationship between effective magnetization and ion flux; n) Effective anisotropy constant K. ef f -FMR Relationship with ion flux; o) at 12 GHz frequency, via Ne + The ion-irradiated sample was at 2.5 × 10⁻⁶. 13 - 2.0×10 14 ions / cm² Fitted ferromagnetic resonance spectra (magnetic field in out-of-plane mode) at various flux levels; p) is at 2.25×10 14 ions / cm² and 2.5×10 14 Fitted ferromagnetic resonance spectra (magnetic field in out-of-plane mode) at flux of ions / cm².

[0036] Figure 2 The comparison of magnetic properties of raw, irradiated, and irradiated + voltage-treated samples is shown in: a)-c) hysteresis loops of raw, irradiated, and irradiated + voltage-treated samples, respectively; d) comparison of saturation magnetization of samples with different treatments; e) comparison of effective magnetic anisotropy of samples with different treatments; f) comparison of damping factors of samples with different treatments; g) comparison of zero-field linewidth of samples with different treatments; and h) comparison of anisotropic fields of samples with different treatments.

[0037] Figure 3 The results show the modulation of the magnetic ion response by analyzing the changes in the hysteresis loops of the original and irradiated samples after applying a series of positive and negative voltages, characterized by the anomalous Hall effect (AHE).

[0038] Figure 4 The diagram illustrates the effective magnetic ion response mechanism in the Ta / CoFeB / Pt / MgO / HfO2 system, showing ion migration and interface state changes under different voltages.

[0039] Figure 5 The figure shows the ion fluence-dependent magnetic ion response spectrum, illustrating the variation of the maximum slope of the AHE hysteresis loop with voltage at different fluences.

[0040] Figure 6 For: the activation voltage and time required to achieve a specific magnetic ion response (e.g., a 10% change in the maximum slope of the AHE hysteresis loop) under different ion energies and fluxes.

[0041] Figure 7 The following is a simulation of the depth distribution and effective concentration range (the proportion of regions with atomic concentration > 0.01%) of each element after neon ion irradiation using the binary approximate collision Monte Carlo method SDTrimSP, demonstrating the atomic mixing at the interface.

[0042] Figure 8 For: Displacement per atom (DPA) profile dependent on ion fluence and sum of DPA for each element.

[0043] Figure 9 γDPA is the rate of change of DPA for each element under different ion energies.

[0044] Figure 10 The relationship between effective magnetic anisotropy Keff, saturation magnetization Ms, maximum slope of the AHE hysteresis loop, activation time, voltage, and the total DPA in the CoFeB layer is given.

[0045] Figure 11 The full width at half maximum (FWHM) of the atomic distribution of each element is measured based on HAADF-STEM EDX.

[0046] Figure 12 For: a) the effective range of interdiffusion between metal-oxide interfaces (Co-O, Fe-O, Mg-O, Hf-O); b) the effective range of interdiffusion between metal-metal interfaces (Co-Fe, Co-Ta, Fe-Ta, Ta-Mg).

[0047] Figure 13 Here are the HAADF-STEM and EDX compositional characterization images of the original sample (a), the irradiated sample (b), and the irradiated sample with applied voltage (c).

[0048] Figure 14For: This demonstrates the sample at 25 keV Ne + Under irradiation, as the ion implantation dose increased from 2.5 × 10⁻⁶, 13 Up to 2.0×10 14 ions / cm 2 The magnetic evolution under varying conditions, including a) effective magnetization (Meff); b) non-uniform linewidth broadening (ΔH0); and c) the trend of damping factor (α), is shown. The results demonstrate the sample's performance at 5 keV Ne + Under irradiation, as the ion implantation dose increased from 2.5 × 10⁻⁶, 13 Up to 2.0×10 14 ions / cm 2 The magnetic evolution under change includes d) effective magnetization (Meff); e) non-uniform linewidth broadening (ΔH0); and f) the trend of damping factor (α).

[0049] Figure 15 For: at 25 keV Ne + Irradiation, different ion implantation doses (2.5×10⁻⁶) 13 - 2.0×10 14 The magnetic ion response of the sample under the condition of (ions / cm²) after applying negative voltage (a), b) and subsequently positive voltage (c), d) is shown. The purple step curve represents the applied voltage process, and the value of each grid cell corresponds to the maximum slope of the AHE hysteresis loop after the negative / positive voltage is applied, which is used to reflect the intensity and trend of magnetic anisotropy modulation.

[0050] Figure 16 For: at 5 keV Ne + Irradiation, different ion implantation doses (2.5×10⁻⁶) 13 -2.0×10 14 The magnetic ion response of the sample under conditions of (ions / cm²) after the application of negative voltage (a), b) and subsequently positive voltage (c), d) is shown. The purple step curves represent the voltage application process, with each grid cell corresponding to the maximum slope of the AHE hysteresis loop after the application of negative / positive voltage, reflecting the intensity of magnetic anisotropy modulation. Compared with high-energy irradiation, the overall response amplitude is weaker and the optimal dose window is narrower under low-energy irradiation.

[0051] Figure 17 For: Ne + Distribution patterns of DPA (displacement damage) induced by irradiation (25 keV) in depth and elemental dimensions.

[0052] Figure 18 For: Ne +Distribution of DPA (displacement damage) caused by irradiation (5keV) and its element dependence. Detailed Implementation

[0053] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0054] Example 1

[0055] A method for controlling the perpendicular magnetic anisotropy and magnetic ion effect of magnetic thin films using neon ion implantation includes the following steps:

[0056] (1) A Ta / CoFeB / Pt / MgO / HfO2 multilayer film was deposited on the substrate. The specific process is as follows:

[0057] At room temperature, metal layers (CoFeB, Pt, Ta) are deposited by DC sputtering and MgO and HfO2 layers are deposited by RF sputtering on a thermally oxidized silicon wafer substrate to form a multilayer film of Ta (5 nm) / CoFeB (1 nm) / Pt (0.09 nm) / MgO (2 nm) / HfO2 (3 nm).

[0058] The base pressure of the sputtering system is 1×10 -7 mbar, argon pressure maintained at 5 × 10 -3 mbar.

[0059] (2) Neon ion implantation:

[0060] The multilayer film of Ta (5 nm) / CoFeB (1 nm) / Pt (0.09 nm) / MgO (2 nm) / HfO2 (3 nm) prepared in step (1) was placed in an ion implanter (such as the JANNUS Saclay facility) and neon ions were implanted at an ion energy of 15 keV and a flux of 2.5 × 10⁻⁶ kilometres per second. 13 –2.0×10 14 Irradiation injection was performed on the multilayer film at an angle of 7 degrees relative to the normal of the sample surface within the range of ions / cm².

[0061] (3) Magnetic ion effect inducement:

[0062] 1) Fabrication of ionic liquid gated devices: Approximately 10 μL of ionic liquid [EMI]+[TFSI]− is drop-coated onto the surface of a thin film (e.g., 10 mm × 5 mm). The gated region (e.g., 5 mm × 5 mm) is in contact with a glass substrate covered with a 100 nm indium tin oxide (ITO) layer as the top electrode, and the magnetic thin film is grounded.

[0063] 2) Apply voltage and perform anomalous Hall effect (AHE) measurement:

[0064] A: Applying a negative voltage, gradually increasing from -1.0 V to -3.5 V, with each voltage lasting for a period of time (Table 1), we observed an increase in the maximum slope of the AHE hysteresis loop, indicating PMA enhancement, transforming from the initial in-plane magnetic anisotropy (IMA) to PMA;

[0065] B: Continue applying a negative voltage to -4.0 V (Table 1), and observe that the maximum slope of the AHE hysteresis loop decreases, PMA is suppressed, and it transforms into IMA;

[0066] C: A positive voltage was then applied, gradually increasing from +1.0 V to +4.5 V, with each voltage lasting for a period of time (Table 1). In the second stage (i.e. after the CoFeB / Ta interface was contaminated), reversible recovery of the PMA state was observed.

[0067] Table 1. Duration of application of different voltages

[0068]

[0069] Note: Taking -2.0V voltage as an example means that after applying a voltage of -2.0V for 60s, the AHE hysteresis curve of the anomalous Hall effect is collected. Then, after applying the voltage for another 180s, the second AHE hysteresis curve of the anomalous Hall effect is collected. Finally, after applying the voltage for 300s, the AHE hysteresis curve of the anomalous Hall effect is collected.

[0070] Characterization and Effects:

[0071] Measurements were taken using a vibrating sample magnetometer (VSM) or ferromagnetic resonance (FMR) to compare the hysteresis loops of unirradiated samples and samples irradiated with different fluxes. The experimental results are as follows: Figure 1 As shown in a)-k);

[0072] Results Analysis: 1.Ne + (15 keV) implantation exhibits a flux-dependent effect on the out-of-plane anisotropy of Ta / CoFeB / Pt / MgO / HfO2, initially enhancing then degrading it: at moderate flux levels (~7.5 × 10⁻⁶), the effect is significantly reduced. 13The optimal PMA enhancement and considerable Ms gain are obtained near the ions / cm² range; further increases in flux are weakened by excessive damage and intermixing of the interface / volume.

[0073] 2. Even at the optimal point, the system is still "close to but not fully" achieving full PMA (Meff≈0, and the hysteresis loop is still slightly tilted), which is related to the field limit and local inhomogeneity of the ultrathin layer.

[0074] 3. This optimal flux range also corresponds to the "magnetic / structure pre-tuning state" of the subsequent voltage-driven magnetic ion effect (maximum AHE slope, lowest activation voltage / time), indicating that ion implantation opens an efficiency window for magneto-ionics through mild interfacial mixing and crystallization regulation.

[0075] In the experimental results shown in Figure 1 (a–k), the regular evolution of the hysteresis loop and its corresponding magnetic parameters with the change in Ne+ ion implantation flux can be clearly observed. At low flux levels (approximately 2.5 × 10⁻⁶), the hysteresis loop and its corresponding magnetic parameters show a regular evolution. 13 Up to 5.0×10 13 (ions / cm²), the hysteresis loop of the sample still exhibits obvious in-plane anisotropy. As the flux gradually increases, the squareness of the out-of-plane hysteresis loop begins to increase, and the coercive field also increases accordingly, indicating that the magnetic anisotropy of the system has undergone an initial transformation. This stage can be regarded as the initial stage of the transition from in-plane anisotropy to out-of-plane anisotropy, and its physical nature is related to local interfacial mixing and local structural adjustment.

[0076] The change in the hysteresis loop is most significant when the flux reaches a moderate level (approximately 7.5 × 10¹³ ions / cm²). At this point, the sample exhibits a nearly square out-of-plane hysteresis loop, while the FMR resonance field shifts significantly towards the lower field direction. This indicates that the effective anisotropy (Keff) and effective anisotropy field (Meff) reach their peaks near this flux. This characteristic signifies that the easy axis of magnetization has shifted from in-plane to out-of-plane, and the perpendicular magnetic anisotropy (PMA) of the system is optimally enhanced in this flux range. In other words, this flux can be considered the "optimal enhancement point," not only maximizing the PMA but also providing an ideal magnetic foundation state for subsequent voltage-controlled magnetism (i.e., the magneto-ion effect).

[0077] When the injection volume is further increased to 1.0 × 10 14 Up to 1.5×10 14At ions / cm², the hysteresis loop of the system begins to change. Although the overall out-of-plane component remains strong, the saturation magnetization in the in-plane and out-of-plane directions no longer coincides, reflecting stronger local inhomogeneity and local magnetic differences within the system. The change in Ms also reflects this trend: it first rises and then plateaus, indicating that the magnetic enhancement effect initially caused by interface improvement has gradually saturated at this stage.

[0078] When the betting volume continues to increase to the highest range (approximately 2.25 × 10), 14 When the flux is ions / cm² or higher, the hysteresis loop and FMR signal of the system show significant degradation. At this point, the squareness of the loop decreases significantly, and the FMR resonance peak becomes blurred or even unresolvable at 12 GHz, indicating that the non-uniformity, damping effect, and interfacial structural damage of the system are very serious, and the PMA begins to weaken significantly. This phenomenon is closely related to excessive intermixing caused by high flux, destruction of the CoFeB layer structure, and Co / Fe–Ta alloying.

[0079] Overall, Figure 1 (a–k) reflects a typical injection dependency relationship of "first enhancing, then degrading". + Ion implantation at low to medium flux levels enhances interfacial anisotropy through mild interfacial mixing and localized crystallization, thereby improving the magnetic heterostructure (PMA) and pushing the Meeff (Mee) towards a critical state close to 0 Oe. This state corresponds precisely to the high-sensitivity response region of voltage-controlled magnetometry. However, when the flux exceeds the optimal range, excessive mixing and structural degradation weaken the PMA and lead to overall magnetic degradation. Therefore, the optimal flux range (approximately 7.5 × 10¹³ ions / cm²) not only achieves near-full PMA magnetic enhancement but also represents an ideal "magnetic pre-tuning state" for efficient magnetic ion control. This result has significant guiding implications for optimizing magnetic heterostructures and achieving low-power magnetoelectric control.

[0080] The variations in saturation magnetization (all saturation magnetizations were calculated based on the nominal thickness (1 nm) of CoFeB), effective magnetization, and effective anisotropy constant Keff of samples irradiated with different fluxes were measured. The experimental results are as follows: Figure 1 As shown in l)-n).

[0081] Results analysis: l) It shows that the saturation magnetization Ms increases with Ne + The betting volume shows a trend of "first rise - plateau - decline": around 5.0 × 10 13 The concentration of ions / cm² reached a peak around 1.25 × 10⁻⁶, and then remained at 1.25 × 10⁻⁶. 14 –1.75×10 14 ions / cm2 The range remained relatively flat; if the betting volume continued to rise to a higher level, it would begin to fall back.

[0082] m) indicates that the effective magnetization Meff increases with flux to approximately 7.5 × 10⁻⁶. 13 ions / cm 2 The value decreased significantly to near zero and fluctuated within a narrow range of -250 to +250 Oe, which corresponds to the system being pushed towards the critical region of "near full PMA", but true full PMA was not achieved at any injection level.

[0083] n) The effective anisotropic energy Keff increases monotonically with flux to approximately 1.5 × 10⁻⁶. 14 ions / cm 2 After that, it fluctuates, consistent with the changes in Ms and Meff, reflecting the optimal enhancement of PMA at medium injection levels and the deterioration trend at higher injection levels.

[0084] The mechanism of the above behavior stems from the competition of multiple factors: at low to medium flux levels, the irradiation-induced crystallization of CoFeB and the enhanced hybridization of Fe(Co)3d–O2p improve Ms and interfacial PMA; at higher flux levels, the excessive diffusion of CoFe to MgO and the paramagnetic alloying of Co / Fe–Ta lead to magnetic dilution and weaken the interfacial anisotropy, causing Ms and Keff to fall back and Meff to be difficult to stabilize as negative.

[0085] Conclusion: Ne + There is an optimal window for irradiation with a moderate flux, which can push the sample to the critical operating point of Meff≈0 and maximize Keff, while achieving a high Ms. Beyond this window, non-homogeneity and magnetic dilution take precedence, and PMA and Ms degrade simultaneously. Therefore, higher flux is not always better, and complete PMA has not been achieved within this working range.

[0086] The performance of the device at 12 GHz was measured via Ne + The ion-irradiated sample was at 2.5 × 10⁻⁶. 13 -2.0×10 14 ions / cm 2 Fitted ferromagnetic resonance spectra (magnetic field in out-of-plane mode) under various flux levels, experimental results are as follows: Figure 1 As shown in o);

[0087] Results Analysis: o) This demonstrates the different flux levels of Ne at 12 GHz, out-of-plane FMR mode. +The resonance spectrum of the irradiated sample. As the flux increased from 2.5 × 10¹³ ions / cm² to 7.5 × 10¹³ ions / cm², the resonance peak position shifted significantly towards the lower magnetic field, indicating a decrease in the effective magnetization of the system and an increase in the vertical anisotropy of the interface. When the flux continued to increase to 2.0 × 10¹³ ions / cm², the resonance peak position shifted significantly towards the lower magnetic field, indicating a decrease in the effective magnetization of the system and an increase in the vertical anisotropy of the interface. 14 When the ions / cm² value is increased, the resonance peak becomes wider and the symmetry deteriorates, indicating that the internal inhomogeneity of the system is enhanced, the magnetic damping and broadening increase significantly, and the FMR signal quality decreases.

[0088] This trend is consistent with the changes in the hysteresis loop and Keff. At medium flux, the PMA enhancement is most pronounced, corresponding to the lowest resonance field and clear spectral lines; at high flux, the structure is disrupted, the magnetism becomes inhomogeneous, the PMA begins to degenerate, and the resonance peaks are difficult to fit clearly.

[0089] Therefore, the results of o) indicate that the optimal magnetic modulation window is located in the middle Ne + Within the flux range, the system approaches the critical state of Meff≈0, where PMA is strongest; however, high flux leads to performance degradation, which is not conducive to achieving stable and strong-response voltage regulation.

[0090] The value was measured at 2.25 × 10⁻⁶. 14 ions / cm 2 and 2.5×10 14 ions / cm 2 The fitted ferromagnetic resonance spectrum (magnetic field in out-of-plane mode) under the flux is shown in the experimental results. Figure 1 As shown in p), no ferromagnetic resonance spectrum was observed, indicating increased inhomogeneity of the system.

[0091] Results analysis: p) indicates that at 2.25 × 10 14 and 2.5×10 14 At high fluxes of ions / cm², the 12 GHz out-of-plane FMR signal is completely invisible. This phenomenon typically indicates a highly non-uniform magnetic field in the system: local anisotropic fields and saturation magnetization are spatially discrete, causing resonance conditions to be "broadened" beyond the measurement window or spectral lines to become indistinguishable, thus preventing the formation of fittable resonance peaks. Combining the definitions of linewidth and broadening parameters in the text, excessive irradiation significantly enhances non-uniform broadening (ΔH0) and increases effective damping, causing the resonance, which was originally clear at moderate fluxes, to be "submerged" by defects, mixing, and magnetic dilution. This aligns with the authors' attribution of this to increased non-uniformity in the figure captions.

[0092] The conclusion is that high-volume Ne +Irradiation pushes the sample from a moderate-flux "resolvable, near-critical PMA operating point" into a highly inhomogeneous and broadened unstable region; the disappearance of the FMR response indicates a significant degradation in magnetic quality. Therefore, the optimized process window should be limited to ≤2.0 × 10⁻⁶. 14 Within the range of ions / cm², to maintain observable and fittable FMR characteristics, further increasing the flux does not bring performance gains, but rather undermines the effectiveness of magnetic properties and subsequent voltage regulation.

[0093] In summary, the irradiated sample exhibits significant PMA enhancement, such as a significant increase in the effective magnetic anisotropy constant Keff-FMR compared to the original sample. The hysteresis loop changes from an in-plane easy axis to a nearly perpendicular easy axis. Figure 1 ).

[0094] Further, this application aims to study the original sample, after exposure to 15 keV and 7.5 × 10¹³ ions / cm² Ne + Changes in the magnetic properties of irradiated samples and samples subjected to irradiation followed by grid voltage treatment. Figure 2 In Figures a–c), VSM measurements were performed on samples in three states in the in-plane (black) and out-of-plane (red) directions, respectively. A) is the original sample; b) is the irradiated sample; c) is the irradiated sample after voltage processing; d) saturation magnetization Ms extracted from the VSM hysteresis loop (black for in-plane and red for out-of-plane); e) effective magnetization Meff; f) effective magnetic anisotropy constant Keff; g) Gilbert damping coefficient α; h) non-uniform linewidth broadening ∆H0 measured from FMR.

[0095] Results analysis:

[0096] Figure 2 In Figure a), the hysteresis loop of the original sample is shown. The hysteresis loop of the original sample is distinctly square in the in-plane direction, while it is more inclined in the out-of-plane direction, indicating a large saturation field. This characteristic suggests that the original sample is dominated by in-plane magnetic anisotropy, with weak perpendicular magnetic anisotropy. Ms is significantly higher in the in-plane direction than in the out-of-plane direction, indicating that the system is in a typical IMA (in-plane magnetic anisotropy) state.

[0097] Figure 2 (b) shows the hysteresis loop of the sample after irradiation; after passing through a 15 keV, 7.5 × 10¹³ ions / cm² Ne... +After irradiation, the shape of the hysteresis loops changed significantly. The squareness of the out-of-plane loops increased markedly, and the coercive field grew, while the in-plane loops became relatively tilted. This indicates that through moderate irradiation, the magnetic easy axis of the system shifted from the in-plane direction to a direction closer to the out-of-plane, effectively enhancing the perpendicular magnetic anisotropy. Ms also increased slightly, indicating that irradiation promoted local crystallization and interface improvement of the CoFeB layer.

[0098] Figure 2 c) shows the hysteresis loop of the irradiated sample after grid voltage modulation, illustrating the loop characteristics after positive and negative voltage modulation. The out-of-plane hysteresis loop becomes more squared, the in-plane hysteresis loop saturation field increases, and Ms also undergoes reversible or partially reversible modulation. This indicates that in the ion-preconditioned sample, interfacial ion migration and electric field modulation can effectively alter magnetic anisotropy, achieving voltage-induced PMA changes. This result verifies the synergistic effect of ion pretreatment and voltage control.

[0099] Figure 2 Figure d) shows the comparison of saturation magnetization of samples from different treatment groups; after irradiation, Ms is significantly higher than in the original state; after voltage regulation, Ms shows a controllable change, indicating that the electric field changes the interface environment and affects the effective ferromagnetic volume or magnetic moment distribution.

[0100] Figure 2 Figure e) shows a comparison of the effective magnetization of samples from different treatment groups; the change in Meeff clearly reflects the evolution of the magnetic anisotropy of the system. The original sample has a larger Meeff, indicating that in-plane anisotropy is dominant; after moderate Ne... + After irradiation, Meff decreased significantly, and the system approached the PMA critical state; after voltage regulation, Meff decreased further, exhibiting stronger perpendicular magnetic anisotropy.

[0101] Figure 2 Figure f) shows the comparison of the effective magnetic anisotropy of samples from different treatment groups, which is consistent with the changes in Meeff. After irradiation, Keff significantly increased compared to the original state, and further increased after voltage application. This indicates that ion irradiation creates an ideal initial magnetic state for voltage-driven interface modulation, allowing the energy barrier to be further adjusted through an electric field.

[0102] Figure 2g) and h) in the figure show the changes in the Gilbert damping coefficient α and the non-uniform broadening ΔH0, respectively. The original sample has low broadening and relatively stable damping; after irradiation, α does not change much, but ΔH0 increases slightly, indicating that although the PMA is enhanced, it is accompanied by a certain degree of interface perturbation. After voltage regulation, α is still within an acceptable range, and ΔH0 does not deteriorate significantly, indicating that voltage regulation does not destroy the magnetic properties of the system and maintains a good FMR response.

[0103] In summary, Figure 2 The results clearly reveal the three-step evolution of magnetic anisotropy modulation: the initial state is IMA, followed by appropriate Ne + After irradiation, the magnetomagnetic anisotropy (PMA) enters a critical state, which is further enhanced by voltage regulation. This process is accompanied not only by a moderate increase in Ms, but also by controllable changes in Meff and Keff, while maintaining good magnetic resonance characteristics. In conclusion, the combined strategy of moderate ion irradiation and voltage-driven operation can efficiently achieve engineered tuning of magnetic anisotropy, providing a feasible implementation path for low-power magnetoelectric devices.

[0104] Furthermore, in order to study the magnetic ion response effects of the original sample and the irradiated sample after applying a series of positive and negative voltages, this application describes the hysteresis loops characterized by the anomalous Hall effect (AHE) after applying a series of positive and negative voltages to the original sample and the irradiated sample, as shown below. Figure 3 As shown.

[0105] Results Analysis: Figure 3 shows the evolution of the AHE loop of the sample under different voltages, revealing the triggering and regulation process of the magneto-ionic effect.

[0106] Figures 3a) and 3b) show the AHE responses of the as-grown samples after applying negative and positive voltages, respectively. The curves show no significant changes, indicating that due to the high crystallinity of MgO, the barrier effect of the Pt layer, and the limitation of the CoFeB structural state, oxygen ions cannot migrate, the voltage cannot modulate the magnetic anisotropy, and the samples do not exhibit a magnetic-ionic response.

[0107] Figure 3c) shows the effect of Ne + After irradiation, under a negative voltage ranging from -1.0 V to -3.5 V, the AHE lap gradually changed from IMA to PMA. This Regime I stage corresponds to the moderate oxidation of the CoFeB layer, forming the maximum PMA, which is key to the activation of the magneto-ionic effect.

[0108] (Figure 3d) After experiencing the series of negative voltages described above, the entire system is now in Regime I. Applying a series of positive voltages as shown in Table 1 in this state does not change the magnetic hysteresis curve. Therefore, the Regime I state that the system enters is an irreversible process (that is, positive voltages cannot restore the original magnetic state of the system).

[0109] Figure 3e) shows that when the voltage is further increased to –4.0 V, the PMA weakens or even degenerates into the IMA, corresponding to the Regime II stage. This is because excessive accumulation of oxygen ions at the CoFeB / Ta interface leads to interfacial contamination. Regime II is reversible.

[0110] Figure 3 f) The PMA partially recovers after applying a positive voltage (a series of positive voltages in Table 1), indicating that Regime II can achieve interface decontamination through voltage, thereby realizing reversible control of magnetic anisotropy, while Regime I is irreversible, and Regime II is reversible.

[0111] Overall, Ne + Irradiation activates oxygen migration channels, enabling the sample to transition from a non-responsive state to a two-stage magneto-ionic response: Regime I (irreversible oxidation) and Regime II (reversible interface contamination / decontamination), providing structural and process basis for voltage-controllable magnetic devices.

[0112] Meanwhile, this application analyzes the effective magnetic ion response mechanism in the Ta / CoFeB / Pt / MgO / HfO2 system, demonstrating ion migration and interface state changes under different voltages, as shown in the specific schematic diagram. Figure 4 As shown.

[0113] Phase 1 (IMA → PMA, Figures 4a–b):

[0114] When the negative voltage increases from –1.0 V to –3.5 V, the electric field induces oxygen ions in the top MgO / HfO2 layer to dissociate and migrate downwards to the CoFeB layer. During this process: (1) oxygen ions preferentially accumulate at the grain boundaries of CoFeB or undergo slight bonding with Co / Fe, leading to local oxidation; (2) the introduction of oxygen significantly enhances the hybridization of Fe(Co)-3d and O-2p orbitals, changing the electronic structure of the interface, which is the key physical mechanism for PMA enhancement; (3) previously Ne +The injection causes partial dissolution or diffusion of the Pt layer, weakening the barrier effect and making it easier for oxygen ions to contact CoFeB. Under the combined effect of these three factors, the magnetic anisotropy changes from in-plane to perpendicular, the PMA reaches its maximum, and the corresponding AHE lap becomes more square and its slope increases. At this point, the oxygen ions are in a stable position within the CoFeB grain boundaries and nanocrystalline regions, and subsequent positive voltage is insufficient to completely remove them; therefore, the PMA at this stage is irreversible.

[0115] Phase 2 (PMA → IMA, Figure 4c):

[0116] When the negative voltage is further increased to -4.0 V, some oxygen ions migrate further downwards to the CoFeB / Ta interface and accumulate. Since Ta is easily oxidized, these oxygen ions react with Ta to form a non-magnetic TaOx layer. This interfacial contamination disrupts the electronic structure coupling of CoFeB / Ta, weakens the PMA generated at the CoFeB / MgO interface, and causes the magnetic anisotropy to revert to IMA, corresponding to a decrease in the slope of the AHE hysteresis loop.

[0117] Third stage (interface decontamination and reversible restoration, Figure 4d):

[0118] When a positive voltage of +1.0 V to +4.5 V is applied, oxygen ions are driven away from the CoFeB / Ta interface, the interface returns to the metallic state, and PMA is restored. The second-stage transition is essentially a reversible regulation process of the interfacial oxygen ion concentration; negative voltage achieves contamination, and positive voltage achieves decontamination, similar to a reversible chemical reaction. In contrast, the oxidation-intercalation process in the first stage corresponds to an irreversible phase transition.

[0119] Figure 5 The ion fluence-dependent magnetic ion response spectra are presented, showing the variation of the maximum slope of the AHE hysteresis loop with voltage under different fluences.

[0120] Results Analysis: Figure 5 shows the results of 15 keV Ne + Under irradiation, the sample was subjected to different ion fluxes (2.5 × 10⁻⁶). 13 –1.5×10 14 Magneto-ionic response under conditions of ions / cm². The horizontal axis represents the sample number, and the vertical axis represents the applied voltage step; each cell represents the maximum slope value of the AHE lap, used to characterize the intensity of perpendicular magnetic anisotropy (PMA). Where a) and b) correspond to the negative voltage stage, and c) and d) correspond to the positive voltage stage.

[0121] Figure 5a) PMA response distribution under negative voltage: at low flux (<2.5×10¹³ ions / cm²) and high flux (>1.5×10¹³ ions / cm²) 14 In the region of (ions / cm²), the magnetic ion response is weak; in the medium flux range (approximately 5×10⁻⁶), the response is weaker. 13 –1×10 14 The ions / cm² showed a significant enhancement, exhibiting a typical non-monotonic dependence, indicating the existence of an optimal irradiation intensity window that can effectively improve magnetic anisotropy.

[0122] Figure 5 b) Negative voltage application sequence: The negative voltage step curves (–1.0 to –4.0 V) are shown as purple steps. All samples experienced the same voltage path, confirming that the differences in PMA are due to fluence effects rather than variations in pressure conditions.

[0123] Figure 5 c) PMA response distribution under positive voltage: After the positive voltage is applied, the medium fluence sample still maintains a strong response, but the overall slope value is slightly reduced compared with the negative voltage stage, indicating that some magnetic anisotropy is reversible, corresponding to the oxygen decontamination-recontamination process (Regime II).

[0124] Figure 5 d) Positive voltage application sequence: The positive voltage step curve (+1.0 to +4.5 V) is also marked with purple steps to ensure consistency of the test trajectory. "X" marks indicate measurement points skipped due to high leakage current.

[0125] In summary, the magnetic ion response exhibits a non-monotonic variation with flux: the response weakens at low and high flux levels, while it is strongest at medium flux levels (boosting regime). Moderate Ne + Irradiation can induce mild interfacial mixing and grain regulation, activate oxygen migration channels, and enhance PMA; excessive irradiation, on the other hand, introduces defects and alloying, weakening the magnetoelectric response. Figure 5 quantitatively reveals the optimal irradiation window for the magneto-ion effect, providing experimental evidence for precise interfacial regulation.

[0126] Figure 6 The activation voltage and time required to achieve a specific magnetic ion response (e.g., a 10% change in the maximum slope of the AHE hysteresis loop) are shown for different ion energies and fluxes.

[0127] Results analysis:

[0128] Figure 6c) Maximum slope variation with flux: The slope exhibits a typical non-monotonic variation, peaking around 7.5 × 10¹³ ions / cm², corresponding to the strongest PMA; the response weakens in both low and high flux ranges. The results indicate that the enhancement effect is most significant at 15 keV, representing the ideal energy condition for achieving "optimal interface mixing" and "controllable crystallization".

[0129] Figure 6 d) Activation voltage and activation time: Both reached their minimum values ​​around 7.5 × 10¹³ ions / cm², indicating that oxygen ion migration kinetics were most efficient and the interfacial reaction barrier was lowest within this dose range (boosting regime). With further increases in flux, both activation voltage and time increased, suggesting that excessive irradiation induced damage and diffusion, reducing magnetic ion activity.

[0130] Conclusion: 15 keV Ne + An optimal flux window exists under irradiation, which can achieve the maximum magneto-ionic response and the lowest energy consumption in the Ta / CoFeB / Pt / MgO / HfO2 system.

[0131] Figure 7 The simulation results based on SDTrimSP are shown. 15 keV Ne + Interfacial element distribution and miscibility characteristics of irradiated samples.

[0132] Results analysis: Figure 7 a) The elemental concentration depth distribution shows that “soft” mixing occurs at all three interfaces, especially CoFeB / Ta; the Pt layer shows signs of dissolution and diffusion as the injection rate increases. Figure 7 b) The effective concentration range of each element has expanded overall, except for Pt, which has shrunk, reflecting its active participation in interface evolution.

[0133] The results revealed that moderate mixing promoted the enhancement of magneto-ionic, while excessive mixing led to a decrease in Ms and inhibition of the effect.

[0134] Figure 8 Demonstrated 15 keV Ne + Under irradiation, different ion fluxes (2.5 × 10⁻⁶) 13 –2.0×10 14 The influence of ions / cm² on the internal displacement damage (DPA) of the sample was investigated, revealing the characteristics of energy deposition distribution and interface structure evolution.

[0135] Figure 8a) DPA Depth Distribution: Within the MgO layer, DPA increases monotonically with depth, indicating that energy is mainly deposited in the oxide layer near the surface. In the CoFeB layer, DPA decreases significantly, decaying rapidly near the Ta layer, suggesting limited direct impact damage to the ferromagnetic layer. With increasing flux, the overall DPA level increases linearly, reflecting the cumulative effect of irradiation damage. This distribution reveals that energy is concentrated at the MgO / HfO2 interface and in the upper region of CoFeB, forming a "soft intermixing zone," which lays the foundation for the formation of oxygen migration channels.

[0136] Figure 8 b) Element dependence: The cumulative DPA of each element increases monotonically with the injection volume, and the damage intensity is as follows:

[0137] O > Ta > Mg > Fe > Co > Hf > B > Pt. The high DPA values ​​of oxygen and Ta indicate that they are most prone to displacement and structural rearrangement: the formation of oxygen vacancies enhances the migration ability driven by the electric field, and the reconstruction of the Ta layer helps to establish an adjustable oxygen capture / release interface.

[0138] Conclusion: DPA exhibits a linear relationship with flux amount, showing a gradual accumulation characteristic. Under moderate flux amount, high damage to O and Ta activates controllable defect states and enhances the magneto-ionic effect; excessive flux amount leads to non-uniform damage and alloying, weakening interfacial reversibility.

[0139] Figure 9 Demonstrated at 5 keV, 15 keV, and 25 keV Ne + Under irradiation conditions, the γDPA (DPA change rate) of each element changes with energy. Except for Ta, the γDPA of all elements decreases with increasing energy; the γDPA in the Ta region peaks at 15 keV, indicating that this energy produces the most effective mixing at the CoFeB / Ta interface, corresponding to the magneto-ionic boosting range. 5 keV only forms shallow defects, 25 keV causes dispersed damage, and 15 keV is the optimal energy window.

[0140] Figure 10. Quantitative correlation between radiation damage (DPA) and magnetic and magnetic ion responses. Figure 10 a) Keff and Ms first increase and then decrease with DPA; Figure 10 b) The PMA strength peaks at DPA ≈ 11, corresponding to the optimal structural equilibrium; Figure 10 c) The activation voltage and time are at their minimum within the same range.

[0141] The results revealed 15 keV Ne + An optimal window of functional DPA ≈ 11 exists under irradiation, achieving maximum PMA and the fastest magnetic ion response.

[0142] Figure 11 The cross-sectional structures of Ta / CoFeB / Pt / MgO / HfO2 multilayer films were presented in three states: unirradiated, irradiated with 15 keV Ne⁺ (7.5 × 10¹³ ions / cm²), and after subsequent voltage treatment. The results show that ion irradiation transforms the clearly defined interlayer boundaries into mildly mixed layers, especially with Co and Fe diffusing downwards and mixing with Ta to form transition layers. Voltage treatment primarily induces downward oxygen migration without disrupting the layer structure. Conclusion: Ne⁺ irradiation significantly improves interfacial atomic mixing, creating diffusion channels for voltage-driven oxygen migration, and is a key prerequisite for enhancing the magneto-ionic effect.

[0143] Figure 12 The overlap range of metal-oxygen (Co–O, Fe–O, etc.) and metal-metal (Co–Ta, Fe–Ta, etc.) element pairs was analyzed. Ion irradiation significantly expanded the metal-metal overlap region, especially the mixing of Co / Fe and Ta, while the increase in metal-oxygen overlap was relatively limited. Conclusion: Ion irradiation-induced metal-metal interface alloying is the main feature of structural regulation, providing a pathway for subsequent oxygen migration and magnetic modulation.

[0144] Figure 13 shows the system at 15 keV Ne + The interfacial structure evolution of Ta / CoFeB / Pt / MgO / HfO2 multilayer films under the combined effects of irradiation and voltage driving was investigated. The results show that irradiation first induces controlled "soft" atom mixing and mild lattice perturbation at the CoFeB / Ta interface, providing the structural prerequisite for the formation of oxygen ion migration channels. Subsequently, voltage driving promotes the directional migration and re-oxidation of oxygen at the interface, leading to reversible reconstruction of the interfacial structure.

[0145] Quantitative analysis using FWHM revealed a significant expansion in the distribution range of Fe, Co, and Mg, indicating that irradiation and voltage treatment together transformed the interface from a sharp stratification to a continuous transition layer, enhancing ion migration activity but simultaneously increasing magnetic damping and inhomogeneity. Further analysis of the coexistence region showed that the interdiffusion between Co–Ta and Fe–Ta was the most significant, representing the main active region for magnetic ion reactions; voltage-driven reversible oxidation / deoxidation cycles were achieved at this interface.

[0146] Overall, Figures 11–13 demonstrate the synergistic mechanism of "irradiation activation + voltage regulation": irradiation activates magnetic ion channels through mild interfacial mixing, while voltage driving achieves redox coupling and interfacial reconstruction. This synergistic effect is the microscopic origin of the reversibility and performance improvement of magneto-ionic interfaces, providing a key structural basis for realizing programmable magneto-electric interfaces.

[0147] Example 2

[0148] A method for controlling the perpendicular magnetic anisotropy and magnetic ion effect of magnetic thin films using neon ion implantation includes the following steps:

[0149] (1) A Ta / CoFeB / Pt / MgO / HfO2 multilayer film was deposited on the substrate. The specific process is as follows:

[0150] At room temperature, metal layers (CoFeB, Pt, Ta) were deposited by DC sputtering and MgO and HfO2 layers were deposited by RF sputtering on a thermally oxidized silicon wafer substrate, forming a multilayer film of Ta (5 nm) / CoFeB (1 nm) / Pt (0.09 nm) / MgO (2 nm) / HfO2 (3 nm).

[0151] The base pressure of the sputtering system is 1×10 -7 mbar, argon pressure maintained at 5 × 10 -3 mbar.

[0152] (2) Neon ion implantation:

[0153] The multilayer film of Ta (5 nm) / CoFeB (1 nm) / Pt (0.09 nm) / MgO (2 nm) / HfO2 (3 nm) prepared in step (1) was placed in an ion implanter (JANNUS Saclay facility) and neon ions were implanted at an ion energy of 25 keV and a flux of 2.5 × 10⁻⁶. 13 –2.0×10 14 Irradiation injection of multilayer films was performed at an angle of 2 degrees relative to the normal of the sample surface, with ions / cm².

[0154] (3) Magnetic ion effect inducement:

[0155] 1) Fabrication of ionic liquid gated devices: Approximately 10 μL of ionic liquid [EMI]+[TFSI]− is drop-coated onto the surface of a thin film (e.g., 10 mm × 5 mm). The gated region (e.g., 5 mm × 5 mm) is in contact with a glass substrate covered with a 100 nm indium tin oxide (ITO) layer as the top electrode, and the magnetic thin film is grounded.

[0156] 2) Apply voltage and perform anomalous Hall effect (AHE) measurement:

[0157] A: Applying a negative voltage, gradually increasing from -1.0 V to -3.5 V, with each voltage lasting for a period of time (Table 1), we observed an increase in the maximum slope of the AHE hysteresis loop, indicating PMA enhancement, transforming from the initial in-plane magnetic anisotropy (IMA) to PMA;

[0158] B: Continue applying a negative voltage to -4.0 V (Table 1), and observe that the maximum slope of the AHE hysteresis loop decreases, PMA is suppressed, and it transforms into IMA;

[0159] C: A positive voltage was then applied, gradually increasing from +1.0 V to +4.5 V, with each voltage lasting for a period of time (Table 1). In the second stage (i.e. after the CoFeB / Ta interface was contaminated), reversible recovery of the PMA state was observed.

[0160] Figure 14 (ac) shows 25 keV Ne + The evolution of magnetic properties of Ta / CoFeB / Pt / MgO / HfO2 multilayer films under irradiation with ion dose.

[0161] Figure 14 a) The effective magnetization (Meff) exhibits a typical "decreasing then increasing" trend with dose: it decreases significantly in the low to moderate dose region (approximately 2.5 × 10¹³–7.5 × 10¹³ ions / cm²), indicating a shift in magnetic anisotropy from IMA to PMA, corresponding to the magneto-ionic boosting region; when the dose is further increased to 2.0 × 10¹³, the magnetization decreases further. 4 When the ions / cm² value is reached, Meff increases, indicating that the interface structure is excessively damaged and PMA decreases.

[0162] Figure 14 b) Non-uniform broadening term (ΔH0): The error bar widens significantly at medium and high doses. Although the average value does not change much, it reflects the enhanced magnetic non-uniformity inside the film, which is related to interfacial mixing and oxygen distribution fluctuations. It is a characteristic of the transition from boosting to suppression.

[0163] Figure 14 c) The Gilbert damping coefficient (α) increases monotonically with dose: it increases moderately at medium doses, reflecting interface softening and local dissolution of the Pt layer; it increases significantly at high doses, indicating increased interface roughness and defect density, enhanced spin scattering, and decreased magnetic response efficiency.

[0164] Figure 15 shows the Ta / CoFeB / Pt / MgO / HfO2 multilayer film under different Ne + Ion implantation dose (2.5×10) 13 –2.0×10 14 The evolution of the magneto-ionic response at ions / cm², 25 keV. Each cell represents the maximum slope of the AHE hysteresis loop, i.e., the voltage-induced PMA intensity; the purple step curve corresponds to the negative (ions / cm², 25 keV). Figure 15 b) and positive ( Figure 15 d) Voltage application process.

[0165] Under negative voltage (Figure 15a, b), PMA initially increases and then decreases with increasing dose: low doses show no activation of oxygen migration channels and a weak response; moderate doses (approximately 7.5 × 10¹³ – 1.0 × 10¹³) show a stronger response. 14 Significant enhancement was observed in ions / cm², corresponding to the magneto-ionic boosting window; high doses (≥2.0×10⁻⁶) were observed. 14 The efficiency of ions / cm² decreases due to excessive interface damage.

[0166] Under positive voltage (Fig. 15c, d), the medium-dose sample can partially recover PMA, indicating that oxygen ion migration is reversible; while the high-dose sample has limited recovery, indicating that interface damage leads to a weakening of reversibility.

[0167] Overall, Figure 15 clearly reveals the dose-dependent and reversible characteristics of the magnetonic effect: insufficient migration at low doses, optimal at medium doses, and structural degradation at high doses. This pattern is consistent with the energy window analysis in Figures 5 and 6, indicating the existence of an optimal dose range with "moderate softening" that can achieve maximum magnetonic response and good voltage reversibility, providing a key basis for the design and process optimization of magnetoelectric tunable devices.

[0168] In summary, a moderate ion dose at 25 keV can achieve a balance state with the lowest Meff, moderate non-uniformity, and controllable damping, corresponding to the optimal magneto-ionic boosting range; while high doses lead to excessive interface damage and magnetic property degradation.

[0169] Figure 6e) shows the maximum slope as a function of flux: the slope exhibits a typical non-monotonic variation, peaking in the range of approximately (5-15) × 10¹³ ions / cm², corresponding to the strongest PMA; the response weakens in both low and high flux ranges. The results indicate that the enhancement effect is most significant at 25 keV, which is the ideal energy condition for achieving "optimal interface mixing" and "controllable crystallization".

[0170] Figure 6f) Activation voltage and activation time: Both reach their minimum values ​​near (5-10) × 10¹³ ions / cm², indicating that oxygen ion migration kinetics are most efficient and the interfacial reaction barrier is lowest within this dose range (boosting regime). As the flux increases further, the activation voltage and time rise, indicating that excessive irradiation induces damage and diffusion, reducing magnetic ion activity.

[0171] In summary, 25 keV Ne + Irradiation at a moderate neon ion radiation dose range achieved the maximum PMA response, the lowest threshold voltage, and the fastest activation rate, thus determining the optimal process window for the magnetionic effect.

[0172] Figure 17 shows Ne + Evolution of sample damage distribution and elemental sensitivity under irradiation with dose.

[0173] Figure 17 a) DPA depth distribution shows that damage is mainly concentrated at the MgO / CoFeB interface (approximately 50–60 Å), i.e., the peak region of ion energy deposition. As the dose increases from 2.5 × 10¹³ to 2.0 × 10¹³, the damage decreases. 14 The overall DPA level increased, with a stronger and slightly broadened peak value, indicating continuous accumulation of interfacial damage. At moderate doses, local perturbations can induce interfacial "softening" and the formation of oxygen migration channels, thereby enhancing PMA; while high doses lead to structural disorder and decreased controllability of magnetic response.

[0174] Figure 17 b) Elemental distribution statistics show that O has the highest DPA, making it most prone to displacement and participation in interfacial diffusion; Mg and Ta are next, Co and Fe are at medium levels, and Pt and B are low. This difference indicates that oxygen is the dominant factor in the magneto-ionic effect: at medium doses, the Pt layer is partially weakened, promoting oxygen migration; at high doses, oxygen diffusion becomes uncontrolled, the interface is severely damaged, and the response efficiency decreases. In summary, the Pt / CoFeB interface is the core region for irradiation energy deposition, and the high sensitivity of oxygen determines the magnetoelectric response intensity. Moderate irradiation can enhance interfacial tunability and promote oxygen migration, while excessive irradiation will cause damage accumulation and performance degradation. This result is corroborated by AHE, FMR, and structural characterization results, providing a key basis for optimizing the ion implantation process window.

[0175] Example 3

[0176] A method for controlling the perpendicular magnetic anisotropy and magnetic ion effect of magnetic thin films using neon ion implantation includes the following steps:

[0177] (1) A Ta / CoFeB / Pt / MgO / HfO2 multilayer film was deposited on the substrate. The specific process is as follows:

[0178] At room temperature, metal layers (CoFeB, Pt, Ta) were deposited by DC sputtering and MgO and HfO2 layers were deposited by RF sputtering on a thermally oxidized silicon wafer substrate, forming a multilayer film of Ta (5 nm) / CoFeB (1 nm) / Pt (0.09 nm) / MgO (2 nm) / HfO2 (3 nm).

[0179] The base pressure of the sputtering system is 1×10 -7 mbar, argon pressure maintained at 5 × 10 -3 mbar.

[0180] (2) Neon ion implantation:

[0181] The multilayer film of Ta (5 nm) / CoFeB (1 nm) / Pt (0.09 nm) / MgO (2 nm) / HfO2 (3 nm) prepared in step (1) was placed in an ion implanter (JANNUS Saclay facility) and neon ions were implanted at an ion energy of 5 keV and a flux of 2.5 × 10⁻⁶. 13 –2.0×10 14 Irradiation injection of multilayer films at an angle of 85 degrees relative to the normal of the sample surface, with ions / cm².

[0182] (3) Magnetic ion effect inducement:

[0183] 1) Fabrication of ionic liquid gated devices: Approximately 10 μL of ionic liquid [EMI]+[TFSI]− is drop-coated onto the surface of a thin film (e.g., 10 mm × 5 mm). The gated region (e.g., 5 mm × 5 mm) is in contact with a glass substrate covered with a 100 nm indium tin oxide (ITO) layer as the top electrode, and the magnetic thin film is grounded.

[0184] 2) Apply voltage and perform anomalous Hall effect (AHE) measurement:

[0185] A: Applying a negative voltage, gradually increasing from -1.0 V to -3.5 V, with each voltage lasting for a period of time (Table 1), we observed an increase in the maximum slope of the AHE hysteresis loop, indicating PMA enhancement, transforming from the initial in-plane magnetic anisotropy (IMA) to PMA;

[0186] B: Continue applying a negative voltage to -4.0 V (Table 1), and observe that the maximum slope of the AHE hysteresis loop decreases, PMA is suppressed, and it transforms into IMA;

[0187] C: A positive voltage was then applied, gradually increasing from +1.0 V to +4.5 V, with each voltage lasting for a period of time (Table 1). In the second stage (i.e. after the CoFeB / Ta interface was contaminated), reversible recovery of the PMA state was observed.

[0188] Figure 6 a) Shows the samples at different 5 keV Ne + The changes in magnetic ion response intensity under different ion implantation doses are shown. It can be seen that with increasing implantation dose, the magnetic ion response (characterized by the maximum slope of the AHE hysteresis loop) first increases and then decreases, exhibiting a typical non-monotonic trend. In the low dose range, the interfacial mixing degree is low, oxygen migration channels are not fully established, and the magnetic ion response is weak; as the dose increases to approximately 1.0–1.5 × 10⁻⁶, the magnetic ion response becomes weaker. 14 At ions / cm², the response reaches its peak, where ion irradiation promotes mild interfacial mixing without disrupting the layer structure, thus lowering the oxygen migration barrier; when the dose is further increased to 2.0 × 10¹, the response is significantly enhanced. 4 When the ions / cm², excessive damage and structural defects lead to an increase in interface traps, thereby suppressing magnetoelectric coupling and weakening the response.

[0189] Figure 6 (b) shows the activation conditions for the magnetic ion response at different doses, including changes in activation voltage and activation time. It can be seen that within the moderate dose range (approximately 1.0–1.5 × 10⁻⁶), the response is significantly improved. 14 The sample with the lowest activation voltage and shortest response time (ions / cm²) indicates that the interface state in this range is most favorable for the electric field-driven migration of oxygen ions. In contrast, in the low or high dose range, the voltage requirement increases significantly and the activation time is prolonged, indicating that the interface is either too dense or excessively damaged, which is not conducive to ion transport.

[0190] In summary, Figures a and b together reveal the optimal "enhancement window" for the magnetic ion effect. At moderate doses of Ne... + Under (5 keV) irradiation, moderate interfacial mixing and unobstructed oxygen migration channels resulted in the simultaneous optimization of magnetoelectric response intensity and energy efficiency. This result demonstrates that by rationally controlling the ion irradiation dose, the magneto-ion effect can be precisely regulated at the interfacial level, providing a key technological basis for the development of high-efficiency, low-energy-consumption magnetoelectric storage devices.

[0191] Figure 14 d) Demonstrates the effect at 5 keV Ne +The effective magnetization Meff of the sample under irradiation is shown to change with the ion implantation dose. It can be seen that as the implantation dose increases from 2.5 × 10¹³ to 2.0 × 10¹³, the magnetization increases with the ion implantation dose. 14 The ions / cm² gradually increased, while Meeff showed a monotonically decreasing trend. This indicates that high-dose irradiation caused stronger interfacial mixing and structural perturbation, weakening the perpendicular magnetic anisotropy (PMA), reflecting the characteristic of magnetic weakening caused by ion damage.

[0192] Figure 14 e) The results show the changes in non-uniform linewidth broadening ΔH0 as a function of dose. ΔH0 reflects the magnetic homogeneity of the system, and the results show that it gradually increases with increasing irradiation dose. This trend indicates that as irradiation energy deposition and interface roughening increase, the non-uniformity of the magnetic environment inside the sample increases, generating more regions of magnetic anisotropic fluctuations.

[0193] Figure 14 f) illustrates the variation of the damping factor α. The results show that α reaches its lowest value at a moderate dose (approximately 7.5 × 10¹³ ions / cm²), and then increases significantly at high doses. This indicates that moderate ion irradiation can improve the interface structure and reduce defect scattering, thereby reducing magnetic damping; while excessively high doses cause structural damage, increase the electron-magnetic moment scattering channels, and lead to an increase in α.

[0194] Figure 16 Demonstrated Ta / CoFeB / Pt / MgO / HfO2 multilayer film at 5 keV Ne + Irradiation (ion implantation dose from 2.5 × 10¹³ to 2.0 × 10¹³) 14 The magnetic ion response (AHE) of the sample under negative voltage (a), b) and positive voltage (c), d) conditions (ions / cm²). The color map in the figure represents the maximum slope of the AHE hysteresis loop at each dose, and the purple step curve represents the change in applied voltage. Figure 16 This indicates that under low-energy irradiation of 5 keV, the magnetic ion response also exhibits a non-monotonic "enhancement-saturation-degradation" trend. At moderate doses, the interface achieves moderate mixing and ion migration channels are formed, resulting in the strongest magnetoelectric modulation; however, excessively high doses lead to structural degradation and response attenuation.

[0195] Conclusion: 5 keV Ne + There is an optimal dose window under irradiation that can effectively activate the magnetic ion effect. The overall enhancement of low-energy irradiation is slightly smaller than that of 25 keV, indicating that the coordinated regulation of energy and dose is crucial for interfacial magnetoelectric coupling.

[0196] Figure 18 shows 5 keV Ne+ The distribution of DPA (displacement damage) under irradiation and its element dependence reveal a direct link between interfacial structural evolution and the enhancement of the magneto-ionic effect.

[0197] Figure 18 a) The DPA depth distribution shows that the damage peak is concentrated at approximately 50–60 Å, corresponding to the Pt / CoFeB interface, which is the main region for ion energy deposition. As the dose increases from 2.5 × 10¹³ to 2.0 × 10¹³, the damage peak decreases. 14 The overall DPA level increased, with enhanced peak value and slight broadening. At moderate doses, the interfacial structure was moderately modulated, which was conducive to the formation of oxygen migration channels; however, at high doses, it caused interfacial disorder and degradation of magnetic response.

[0198] Figure 18 b) Element-dependent analysis showed that O had the highest DPA, making it most susceptible to displacement and migration; Mg and Ta were next, Co and Fe were at moderate levels, while Pt and B were low, indicating that the Pt layer still had some blocking effect. This distribution suggests that oxygen is the main driving force of the magneto-ionic effect: at moderate doses, the "softening" of the Pt layer promotes oxygen migration, while high doses induce disordered oxygen diffusion and interface degradation. Overall, DPA damage is mainly concentrated at the Pt / CoFeB interface, and the high sensitivity of oxygen determines the intensity of the magnetoelectric response. Moderate irradiation can achieve controllable modulation of the interface and enhancement of the magneto-ionic effect, while excessive irradiation leads to structural degradation and suppression of the effect.

[0199] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0200] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for controlling the perpendicular magnetic anisotropy of a magnetic thin film using neon ion implantation, characterized in that, Includes the following steps: (1) Deposit a Ta / CoFeB / Pt / MgO / HfO2 multilayer film on the substrate; (2) Neon ions were used at an energy of 5-25 keV and a flux of 2.5 × 10⁻⁶. 13 –2.0×10 14 Irradiation implantation of the multilayer film described in step (1) under the condition of ions / cm² induces soft atom mixing, Pt layer dissolution and CoFeB layer crystal structure change at the CoFeB / Pt / MgO interface and CoFeB / Ta interface, thereby achieving the enhancement of vertical magnetic anisotropy.

2. The method according to claim 1, characterized in that, The neon ion implantation angle is 2–90° relative to the sample surface normal.

3. The method according to claim 1 or 2, characterized in that, The neon ion flux was 7.5 × 10¹³ ions / cm², the energy was 15 keV, and the injection angle was 7° relative to the sample surface normal.

4. A method for inducing and controlling magnetic ion effects in magnetic thin films, characterized in that, Includes the following steps: (1) Pretreatment of Ta / CoFeB / Pt / MgO / HfO2 multilayer film by neon ion implantation as described in any one of claims 1–3; (2) Apply voltage to the injected sample to achieve multi-stage reversible control of magnetic anisotropy by driving oxygen ion migration.

5. The method according to claim 4, characterized in that, The voltage regulation process is as follows: (1) Apply a negative voltage from -1.0 V to -3.5 V to achieve the transformation from in-plane magnetic anisotropy to maximum perpendicular magnetic anisotropy; (2) Based on step (1), continue to apply a negative voltage to -4.0 V to realize the transformation from perpendicular magnetic anisotropy to in-plane magnetic anisotropy; (3) Based on step (2), apply a positive voltage from +1.0 V to +4.5 V to restore the perpendicular magnetic anisotropy.

6. The method according to claim 5, characterized in that, The reversible transformation between vertical magnetic anisotropy and in-plane magnetic anisotropy in steps (2) and (3) is achieved through the contamination and decontamination mechanism of the CoFeB / Ta interface.

7. A magnetic structure, characterized in that, The multilayer film comprising a Ta / CoFeB / Pt / MgO / HfO2 multilayer film treated by any one of claims 1–3, the multilayer film exhibiting enhanced vertical magnetic anisotropy and / or voltage-tunable magnetic ionization.

8. The application of the magnetic structure according to claim 7 in magnetic storage devices or magnetic logic devices.

9. A magnetic random access memory, characterized in that, It includes the magnetic structure as described in claim 7 as a free layer.

Images