Article www.acsnano.org Regulating Oxygen Ion Transport at the Nanoscale to Enable Highly Cyclable Magneto-Ionic Control of Magnetism Zhengwei Tan, Zheng Ma, Laura Fuentes, Maciej Oskar Liedke, Maik Butterling, Ahmed G. Attallah, Eric Hirschmann, Andreas Wagner, Llibertat Abad, Nieves Casañ-Pastor, Aitor F. Lopeandia, Enric Menéndez,* and Jordi Sort* See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles. Cite This: ACS Nano 2023, 17, 6973−6984 Read Online ACCESS * Downloaded via 158.109.58.100 on October 31, 2023 at 12:22:48 (UTC). Metrics & More Article Recommendations sı Supporting Information ABSTRACT: Magneto-ionics refers to the control of magnetic properties of materials through voltage-driven ion motion. To generate effective electric fields, either solid or liquid electro- lytes are utilized, which also serve as ion reservoirs. Thin solid electrolytes have difficulties in (i) withstanding high electric fields without electric pinholes and (ii) maintaining stable ion transport during long-term actuation. In turn, the use of liquid electrolytes can result in poor cyclability, thus limiting their applicability. Here we propose a nanoscale-engineered magne- to-ionic architecture (comprising a thin solid electrolyte in contact with a liquid electrolyte) that drastically enhances cyclability while preserving sufficiently high electric fields to trigger ion motion. Specifically, we show that the insertion of a highly nanostructured (amorphous-like) Ta layer (with suitable thickness and electric resistivity) between a magneto-ionic target material (i.e., Co3O4) and the liquid electrolyte increases magneto-ionic cyclability from <30 cycles (when no Ta is inserted) to more than 800 cycles. Transmission electron microscopy together with variable energy positron annihilation spectroscopy reveals the crucial role of the generated TaOx interlayer as a solid electrolyte (i.e., ionic conductor) that improves magneto-ionic endurance by proper tuning of the types of voltage-driven structural defects. The Ta layer is very effective in trapping oxygen and hindering O2− ions from moving into the liquid electrolyte, thus keeping O2− motion mainly restricted between Co3O4 and Ta when voltage of alternating polarity is applied. We demonstrate that this approach provides a suitable strategy to boost magneto-ionics by combining the benefits of solid and liquid electrolytes in a synergetic manner. KEYWORDS: magneto-electricity, voltage control of magnetism, magneto-ionics, transition metal oxide, ion diffusion W ith the advent of global phenomena such as the a whole area of research referred to as voltage control of Internet of Things, artificial intelligence, machine magnetism (VCM). learning, or Big Data, the demand for highly Magneto-ionics refers to a particular mechanism for VCM in functional and energy-efficient miniaturized microelectronic which voltage-driven ion transport (of, e.g., O2−,11−15 Li+,16 F−,17 H+,18−21 or N3−22−26 species) leads to a large and components is growing exponentially.1,2 Spintronic systems3,4 controllable modulation of magnetism without the need of operated using electric currents through spin-transfer torque5,6 strain transfer. This is different from voltage-controlled strain- or spin−orbit torque7,8 effects are becoming key elements for mediated multiferroic heterostructures, which are less con- next-generation nanoelectronics with enhanced memory and information processing capabilities. However, magnetization Received: February 6, 2023 switching using electric current involves an undesirable Joule Accepted: March 21, 2023 heating effect, which is detrimental to energy efficiency.9,10 An Published: March 27, 2023 interesting alternative is to modulate magnetic properties by applying electric fields instead of electric current, thus minimizing power dissipation. This has rapidly evolved into © 2023 The Authors. Published by American Chemical Society https://doi.org/10.1021/acsnano.3c01105 6973 ACS Nano 2023, 17, 6973−6984 ACS Nano www.acsnano.org Article venient for spintronics because repeated voltage actuation can thicknesses, difficulties arise to withstand high electric fields lead to mechanical fatigue and eventual device failure. without electric pinholes. Moreover, thin solid electrolytes Archetypical magneto-ionic structures comprise a gate electro- offer a limited ion buffering capability (i.e., they easily become lyte (either solid or liquid)27,28 in contact with a ferromagnetic saturated with ions and cannot sustain stable ion transport (FM) target material. Voltage is applied across the electrolyte, during long-term operation, especially when the magneto-ionic using the FM film as a working electrode and causing a voltage layer is thicker than the solid electrolyte layer). Liquid polarity-dependent insertion/removal of ions into/from the electrolytes are convenient for other magnetoelectric applica- target material. In this way, magnetic properties, such as tions, such as magnetophoresis/microfluidics41 or to emulate saturation magnetization,11,22 magnetic anisotropy and co- neuromorphic functionalities (the brain operates in a liquid ercivity,12,18 exchange bias field,13,29 or skyrmion generation/ environment).15,26 Owing to the formation of the “electric suppression,20 among others, can be reversibly controlled, double layer”, whose thickness is <1 nm, liquid electrolytes are making magneto-ionic materials highly promising for ultra-low- able to generate ultralarge electric fields (hundreds of MV power magnetic devices. An extreme case is when voltage cm−1) at the interface between the liquid and the target induces a complete reversible transition between FM and magneto-ionic layer, without electric pinholes.27 Liquid nonmagnetic states, leading to voltage-driven ON−OFF electrolytes may also be good ion reservoirs.11,14,22 However, switching of ferromagnetism. This has been reported in a when voltage is applied, the mobile ions released from the number of magneto-ionic systems, such as Co3O4,11,14,30 Li+- target layer into the liquid can travel long distances toward the intercalated α-Fe2O3,31 F−-intercalated La2−2xSr1+2xMn2O7,17 counter electrode, making their reintroduction into the CoN,26,28 CoMnN,25 or α-Co(OH)2.32 ON−OFF switching of magneto-ionic layer (with voltage of opposite polarity) the ferromagnetic state has been also induced by electrostatic difficult. In addition, at the counter electrode, if sufficiently surface charging in FeS2.33 high voltage is applied, the dissolved ions (e.g., O2−, H+, or In most magneto-ionic systems, the source of the moving N3−) can transform into the corresponding oxygen, hydrogen, ions is electrolytes (e.g., GdOx, HfOx, propylene carbonate with or nitrogen gases and be released to the atmosphere in the dissolved LiPF6 or KI) that are in direct contact with pristine form of bubbles. These effects are difficult to control and are FM or ferrimagnetic layers (e.g., Co, Fe, Fe2O3), whose highly detrimental for magneto-ionic reversibility and endur- properties are manipulated with voltage.27,34−36 Magneto-ionic ance. Restricting ion transport within the magneto-ionic layers systems based on O2− insertion from an external electrolyte to could avoid these problems and is expected to improve the a target FM material often suffer from slow dynamics at room magneto-ionic cyclability. temperature and irreversible compositional/structural changes Here, we propose an improved nanoscale-engineered in the FM phase, eventually leading to degradation and limited magneto-ionic structure that results from inserting an cyclability.12 In addition, for relatively thick films, it is amorphous Ta layer (which gets spontaneously passivated in challenging to achieve a fully OFF magnetic state with voltage air) between the Co3O4 film (magneto-ionic target material) due to the limited penetration of the external ions toward the and propylene carbonate, PC (liquid electrolyte). Upon interior of the FM target layers.37 Recently, smaller ions (e.g., application of negative voltage, O2− ions migrate from H+, F−, or Li+) have been introduced to achieve faster and Co3O4 to Ta, promoting the formation of TaOx (Ta is a more cyclable voltage-induced manipulation of magnetic good oxygen getter), which acts as a thin solid electrolyte with properties. Fast, reversible, and cyclable tuning of perpendic- good ionic conductivity. This architecture preserves sufficiently ular magnetic anisotropy,18 Dzyaloshinskii−Moriya interac- high electric fields to trigger ion motion while allowing to tion,38,39 or ferrimagnetic spin textures40 has been achieved repeatedly induce ON−OFF switching of ferromagnetism at through chemisorbed O2− or H+ ion species. However, systems room temperature. By optimizing the thickness of the Ta layer, relying on H+ are either sensitive to environmental conditions a drastic increase of cyclability is achieved in Co3O4/Ta (25 (e.g., humidity) or restricted by their incompatibility with nm)/PC (>800 cycles) compared to Co3O4/PC with no Ta traditional complementary metal-oxide semiconductor insertion (<30 cycles). The Ta layer hinders oxygen ions from (CMOS)-based devices (i.e., standard fabrication processes entering the liquid electrolyte and allows for O2− redistribution for semiconductor devices, such as metal oxide field-effect inside the Co3O4 and Ta layers, as assessed by high-angle transistors).28 An alternative approach is to use structural annular dark-field scanning transmission electron microscopy oxygen or nitrogen, self-contained in the target materials, as (HAADF-STEM) and electron energy loss spectroscopy the source of the moving ions. Examples of such materials are (EELS). Positron annihilation spectroscopy is used to precisely Co3O4,11 CoN,22,26 or CoMnN25 films. These target materials, characterize the defect structure in Ta, which strongly which are CMOS compatible, exhibit an initially fully OFF contributes to the O2− ion transport. An increase of the defect (i.e., paramagnetic) state and provide “ready-prepared” lattice size during biasing is observed. The reduced electric sites for ion diffusion, allowing for net magneto-ionic conductivity of the passivated amorphous Ta layer is important generation of ferromagnetism by voltage-triggered O2− or to allow the penetration of the electric field inside the Co3O4 N3− ion motion from the films toward a neighboring layer and the concomitant O2− ion transport. Since O2− electrolyte. Unfortunately, achieving a high magneto-ionic diffusion becomes mainly restricted within the Co3O4/TaOx cyclability in these materials (i.e., removing and reinserting the structure (instead of O2− being released to PC), this greatly O2−/N3− ions many times by switching voltage polarity) improves the efficiency of ion transport in the material, remains a challenge. resulting in largely enhanced magneto-ionic cyclability. To induce magneto-ionics, either solid or liquid electrolytes can be utilized. Solid electrolytes (with ultrathin dielectric RESULTS AND DISCUSSION layers) are preferred for solid-state spintronics. Ultrathin solid The basic building block of the investigated magneto-ionic dielectric layers are needed to induce sufficiently large electric system is a 40 nm thick Co3O4 film grown by DC reactive fields under moderate applied voltages. However, at such small sputtering at room temperature onto Cu (60 nm)/Ti (20 6974 https://doi.org/10.1021/acsnano.3c01105 ACS Nano 2023, 17, 6973−6984 ACS Nano www.acsnano.org Article Figure 1. Structural and compositional characterization of as-deposited films. (a) HAADF-STEM micrograph of an as-grown sample. (b) EELS Co (in red), O (in blue), and Ta (in yellow) elemental mappings of the area marked with an orange rectangle in (a). (c) θ/2θ XRD patterns of the as-prepared samples. (d, e) High-resolution TEM images of the cross section of the as-deposited Co3O4 and Ta films, respectively. The inset shows the fast Fourier transform of the areas marked with red squares. In panel (e), the “1” spot corresponds to an interplanar distance of 0.248 nm and is consistent with (311) Co3O4 (PDF 00-009-0418) or (111) CoO (PDF 00-001-1227) interplanar distances (0.244 and 0.245 nm, respectively), while spot “2” corresponds to an interplanar distance of 0.225 nm and is unambiguously ascribed to the (200) CoO (PDF 00-001-1227) interplanar distance (0.212 nm). The “3” spot corresponds to an interplanar distance of 0.448 nm, which is consistent with (311) Co3O4 (0.467 nm, PDF 00-009-0418). nm)/[100]Si (725 μm) substrates. The Co3O4 films were exhibits higher electric resistivity than its crystalline counter- coated with sputtered amorphous Ta protective layers of parts.44 This is beneficial to keep sufficient electric field variable thickness (from 5 to 50 nm), which were left to strength inside the actuated films while voltage is applied, passivate in air (see Experimental Section). Uncoated Co3O4 thereby enabling magneto-ionics. films were also grown as a reference. Figure 1(a) shows a low- To further investigate the structure of the samples, high- magnification HAADF-STEM image of the as-prepared sample resolution transmission electron microscopy (HRTEM) with a 25-nm-thick Ta protective capping layer. Clear images of the cross sections of as-prepared Co3O4 and Ta interfaces between the various layers can be seen, and their layers were taken, as shown in Figure 1(d) and (e), thicknesses are in good agreement with the nominal ones. respectively. The areas marked with red squares were chosen Note that the topmost Pt capping layer was grown only during for fast Fourier transform (FFT) analyses, as shown in the the TEM lamella preparation (i.e., it was not present during insets. In Figure 1(d), the “1” spot corresponds to an magneto-ionic experiments). To investigate the Co, O, and Ta interplanar distance of 0.248 nm and is consistent with distribution, Co, O, and Ta EELS mappings from the area (311) Co3O4 (PDF 00-009-0418) or (111) CoO (PDF 00- marked with an orange rectangle in Figure 1(a) were acquired. 001-1227) interplanar distances (0.244 and 0.245 nm, As shown in Figure 1(b), Co and O elements are respectively), while spot “2” corresponds to an interplanar homogeneously distributed within the Co3O4 films. Addition- distance of 0.225 nm and is unambiguously ascribed to the ally, some O signal is also detected inside the Ta layer, (200) CoO (PDF 00-001-1227) interplanar distance (0.212 particularly on its upper part. This is ascribed to surface self- nm). The “3” spot corresponds to an interplanar distance of passivation of Ta in air, which is known to result in the 0.448 nm, which is consistent with (311) Co3O4 (0.467 nm, formation of a stable TaOx top layer.42,43 Figure 1(c) shows PDF 00-009-0418). This indicates that a mixture of CoO and the θ/2θ X-ray diffraction (XRD) pattern of an as-grown film. Co3O4 phases is plausible in the Co oxide film, in agreement The XRD peak observed at 2θ ≈ 19.1° is consistent with (111) with previously reported results.15 For simplicity, in spite of the Co3O4 (PDF 00-009-0418), whereas that at 2θ ≈ 36.7° could presence of CoO in particular close to the interface, the Co be ascribed to either (111) Co3O4 or (111) CoO (PDF 00- oxide film is labeled Co3O4 throughout the article. Remarkably, 001-1227). Peaks corresponding to the Cu buffer layer are also no spots are observed for Ta, in agreement with its amorphous detected. Conversely, no traces of Ta are evidenced, suggesting nature, as also evidenced by XRD (Figure 1(c)). that Ta is highly nanostructured or even amorphous-like. To induce magneto-ionics, electrolyte gating was performed Growth of amorphous Ta by physical deposition methods has in a capacitor-like configuration (Figure 2(a))14 using a been previously reported,44,45 and it is promoted by the platinum wire as counter electrode and an aprotic, anhydrous existence of ions and particles with high kinetic energy when polar liquid electrolyte composed of propylene carbonate with sputtering at sufficiently high gun powers. Amorphous Ta Na+ and OH− solvated species.27,46,47 When voltage is applied, 6975 https://doi.org/10.1021/acsnano.3c01105 ACS Nano 2023, 17, 6973−6984 ACS Nano www.acsnano.org Article Figure 2. Magneto-ionic characterization of Co3O4 thin films and Co3O4/Ta bilayer films under electrolyte gating. (a) Schematic of the designed structure for electrolyte actuation. (b) Left: Sketch of the formation of electric double layer during voltage actuation; right: photograph of the homemade electrolytic cell used to apply voltage to our system, which consists of an Eppendorf filled with propylene carbonate, our sample as working electrode, and a Pt wire counter electrode. (c) Time evolution of the saturation magnetization, MS, for Co3O4 thin films and Co3O4/Ta bilayer films (with a Ta thickness of 5, 10, 25, and 50 nm) under an applied voltage of −25 V. (d) Estimation of the onset voltage required to trigger magneto-ionics in Co3O4 thin films and Co3O4/Ta (25 nm) bilayer films, as well as the recovery process. (e) Calculated absolute value of the voltage at the Co3O4/Ta interface, |ΔVint|, as a function of the TaOx conductivity, σTaOx, when externally applying |ΔV| = 25 V for different thicknesses of the Ta layer, δTa, and a fixed Co3O4 conductivity, σCo3O4 = 20 S m−1. (f) Calculated interface voltage as a function of σTaOx, for different values of conductivity of σCo3O4 and a fixed Ta thickness (δTa = 50 nm). The regions highlighted with a cyan rectangle in (e) and (f) are those where |ΔVint| would be below the magneto-ionic threshold voltage of the system. a sub-nm-thick electric double layer forms at the electrolyte obtain emu cm−3). As seen in Figure 3, irrespective of the side of the electrolyte/Ta interface, allowing for the generation thickness of the Ta layer (i.e., 0, 5, 10, 25, or 50 nm), all films of a high electric field.47 This electric field is ultimately in the as-grown state show very little ferromagnetic response responsible for driving oxygen ions from the Co3O4 to the Ta/ (<33 emu cm−3, which is equivalent to ≈2.3% the magnet- TaOx layers (as illustrated in Figure 2(b)). Voltage treatments ization of pure FCC-Co48). This small ferromagnetic signal is were performed in situ, while hysteresis loops were recorded at common in sputtered Co3O415 and is likely due to either a room temperature by vibrating sample magnetometry (VSM), small fraction of residual Co clusters that do not become fully with an in-plane applied magnetic field. The total measured oxidized during the sputtering process or substrate contami- magnetic moment of the samples (in emu) was normalized to nation. The deposition of Ta onto Co3O4 has a negligible the area of the sample and the nominal thickness of Co3O4 (to effect on the initial ferromagnetic signal of the Co3O4 films. 6976 https://doi.org/10.1021/acsnano.3c01105 ACS Nano 2023, 17, 6973−6984 ACS Nano www.acsnano.org Article Figure 3. In-plane VSM hysteresis loops (each lasting 30 min) of the as-prepared Co3O4/Ta films (black), the films biased under −25 V (for 30 min in red and for 1 h in blue), and subsequently recovered after applying +25 V for 30 min. The different panels correspond to different Ta thicknesses: (a) 0 nm, (b) 5 nm, (c) 10 nm, (d) 25 nm, (e) 50 nm. Panel (f) corresponds to Co3O4/Pt without a Ta interlayer (with a Pt thickness of 5 nm). Upon negative biasing at −25 V for 30 min or 1 h, clear onset of the oxygen ionic motion, which leaves metallic hysteresis loops are observed. The generated saturation ferromagnetic Co behind.14,15 The obtained steady-state value magnetization (MS) is maximum for the sample without Ta of MS decreases with the Ta thickness, from 401 emu cm−3 to (around 400 emu cm−3, suggesting that 27% of the volume of 19 emu cm−3 (in agreement with the magnetic hysteresis loops Co3O4 gets reduced to Co), and it slightly decreases when a shown in Figure 3(a)−(e)). One important parameter of large thin Ta interlayer (5−10 nm) is inserted between the Co3O4 significance for device applications is the minimum threshold film and the liquid electrolyte. MS further decreases for voltage required to trigger magneto-ionic effects.22,25 The sufficiently thick Ta. Remarkably, magneto-ionic effects when onset voltage for Co3O4 films without and with a 25-nm-thick Pt (instead of Ta) is grown onto Co3O4 are strongly reduced: a Ta capping layer was evaluated by subjecting the films to small hysteresis loop is observed only for 5 nm Pt (compare increasing negative voltage steps of −2 V, until MS started to Figure 3(f) with (b)), and no ferromagnetic response is increase, as shown in Figure 2(d). The results reveal that the obtained for thicker Pt. In all cases, the initial virtually onset voltage is approximately −6 V for both samples, whereas nonmagnetic state can be recovered by applying +25 V for 30 min. a voltage of +8 V leads to complete recovery in both cases, Figure 3 shows that, for all samples, most of the which agrees with previous works on similar systems.11 ferromagnetic response is induced in less than 30 min, since Figure 2(c) and (d) also show that the rate at which MS MS does not further increase when applying voltage for 1 h. increases under voltage application is larger for a smaller Ta The time evolution of MS for all samples under −25 V is layer thickness. This, together with the reduction of the steady- shown in Figure 2(c). An external magnetic field of 10 kOe state MS for thicker Ta, indicates that the effective electric field (i.e., above the anisotropy field of the generated ferromagnetic acting on the Co3O4 layer becomes lower for larger Ta counterpart) was applied during these measurements to ensure thickness. This suggests that, for thicker Ta, there is a more magnetic saturation. In all films, an immediate increase of MS is severe dissipation of space charge density, while O2− is observed in response to the applied ΔV, evidencing a quick migrating from Co3O4 to form TaOx.49 6977 https://doi.org/10.1021/acsnano.3c01105 ACS Nano 2023, 17, 6973−6984 ACS Nano www.acsnano.org Article Figure 4. Magneto-ionic cyclability of (a) Co3O4/Ta (25 nm) bilayer films and (b) the Co3O4 thin films subjected to −25 V/+25 V voltage pulses applied with a periodicity of 80 s. In a first approximation, this effect can be modeled the interface voltage and slowing down the magneto-ionic considering the system as a simple voltage divider. The voltage process. Another interesting effect is that when Co3O4 drop at the Co3O4/Ta interface can be estimated assuming transforms to metallic Co, the conductivity of the magneto- that, during the magneto-ionic process, the Co3O4 and the ionic layer (σCo3O4) increases. This, in turn, has an effect on newly formed TaOx layers act as two resistances connected in ΔVint, which is shown in Figure 2(f). In bulk metallic Co, σCo series. In this case, the voltage at the Co3O4/TaOx interface is can reach values >107 S m−1.52 As plotted in Figure 2(f), when R Vint = R Co+3OR4 V . Considering that R = ρδ/A (where ρ σCo3O4 transforms to Co and σCo3O4 increases, the interface Co3O4 TaOx voltage drastically drops, and ΔVimt easily falls below the is the resistivity of each material, δ is the film thickness, and A threshold voltage. Then any magneto-ionic response will tend is the lateral area of the films), it is straightforward to estimate to stop. As an example, for the particular case of δTa = 50 nm, ΔVint as a function of the layers’ thicknesses and the electric any eventual O2− ion motion triggered while σTaOx is conductivities of Co3O4 and TaOx. The electrical conductivity sufficiently large will tend to stop as soon as σCo3O4 increases of Co3O4, measured by the Van der Pauw method, is above 103 S m−1 (see Figure 2(f)) or σTaOx decreases below 5 S approximately σCo3O4 = 20 S m−1.14 The conductivity of m−1 (Figure 2(e)). It is noteworthy that, although this intuitive TaOx strongly depends on the oxygen content, and it varies by picture is a simplified representation of the reality, it already several orders of magnitude, from 105 S m−1 in amorphous provides some basic understanding of the role of the TaOx metallic Ta44 to 10−6 S m−1 for highly oxidized Ta.50 The layer thickness and resistivity on the induced magneto-ionic presence of oxygen in the naturally passivated Ta (Figure effects. 1(b)) can easily bring ρ to values in the range 10−1−102 S m−1 Notwithstanding the decrease of the steady-state MS with before any voltage is applied.51 Figure 2(e) shows the the deposition of a Ta capping layer (by approximately 25% for evolution of the calculated interface voltage as a function of δTa = 25 nm), the formation of TaOx (which acts as a solid σTaOx for the different Ta film thicknesses. For highly electrolyte) has one very beneficial effect: it drastically conductive Ta (or TaOx) spacer layers, ΔVint becomes enhances magneto-ionic cyclability. This is evidenced in Figure independent of the Ta thickness, which means that any 4, which shows the cyclability results (i.e., repeated increase/ eventual magneto-ionic effect would be independent of the Ta decrease of M) in a Co3O4 thin film and Co3O4/Ta (25 nm) thickness (and ΔVint would be always equal to −25 V). bilayer films upon application of −25 V/+25 V voltage pulses However, in this case, the electric field would be highly with a duration of 80 s. As shown in Figure 4(a) and (b), a screened at the interface and would not penetrate inside Ta very stable and reversible behavior is observed during the first (since, in metals, the electric field is confined within the 10 cycles for both samples. However, the Co3O4/Ta (25 nm) Thomas−Fermi screening length, which is typically <0.5 bilayer sample maintains a very stable cyclability even after 800 nm).27 Thus, for highly metallic interlayers (e.g., noble metals cycles (Figure 4(a)), whereas no sign of magneto-ionic effect is like Pt, as in Figure 3(f)), little magneto-ionic effects are detected for Co3O4 films after 30 cycles (Figure 4(b)). This expected since there is no driving force (no electric field) to demonstrates a significant enhancement of the endurance of induce ion motion inside the metal.24 This is opposite to what the system by the deposition of a Ta overlayer with appropriate happens in semiconductors (e.g., TaOx), where the electric thickness. field will penetrate deeper into the layer. Interestingly, for low The improved cyclability in the Co3O4/Ta (25 nm) bilayer electrically conductive TaOx (i.e., σTaOx < 100 S m−1), ΔVint is ascribed to the role of Ta in allowing diffusion of oxygen becomes clearly lower as the Ta layer thickness increases ions. In this sample, when negative voltage is applied, the O2− (Figure 2(e)). For δTa = 50 nm, if σTaOx is lower than 5 S m−1, ions exiting Co3O4 are captured by Ta, forming a TaOx solid ΔVint eventually drops below the threshold voltage (region electrolyte, instead of being directly released to the liquid highlighted in cyan), meaning that no magneto-ionic effects electrolyte. Ta is a well-known oxygen getter,53 and TaOx is a will be induced in this case (while, for thinner Ta layers, for the good ionic conductor.54 In unprotected Co3O4 films, O2− ions same σTaOx value, ΔVi will still be above the threshold). Note are directly dissolved in the propylene carbonate, and once that during magneto-ionic motion of O2− from Co3O4 to solvated with the PC chains, they can travel long distances TaOx, the values of σTaOx will rapidly decrease, thus reducing toward the positively charged counter electrode (i.e., the Pt 6978 https://doi.org/10.1021/acsnano.3c01105 ACS Nano 2023, 17, 6973−6984 ACS Nano www.acsnano.org Article Figure 5. Compositional characterization of the Co3O4/Ta (25 nm) films gated at −25 V for 1 h. (a) HAADF-STEM and (b, c) elemental EELS mappings corresponding to the area marked with an orange rectangle in the HAADF-STEM image. Cu and Pt layers serve as working electrode and protective capping layer during TEM lamellae preparation, respectively. Co, O, and Ta are represented by red, blue, and yellow colors in the EELS elemental mappings. (d) Depth profile of Co, O, and Ta elements along the dark pink arrow drawn in (a). (e) High-resolution TEM image of the sample treated at −25 V for 1 h. Figure 6. Structural characterization by X-ray diffraction (XRD) and high-resolution transmission electron microscopy (HRTEM). (a) θ/2θ XRD diffraction patterns of Co3O4/Ta (25 nm) films in the as-prepared state and after being treated at −25 V for 1 h. (b) HRTEM images of the cross section of the Co3O4/Ta (25 nm) film after applying −25 V for 1 h. The inset shows the fast Fourier transform (FFT) of the area marked with a red rectangle, and it shows spots with an average interplanar distance of 0.19 nm, which corresponds to (101) hexagonal- closed-packed (HCP) metallic Co. For phase identification, the cards no. PDF 00-005-0727 and PDF 00-009-0418 were taken for Co and Co3O4, respectively. wire), where they eventually form O2 gas bubbles and are transport mechanism, cross-sectional lamellae of a bilayer released to the atmosphere at sufficiently high voltages. This sample (40-nm-thick Co3O4 plus 25-nm-thick Ta overlayer) long-distance transport of ions in the liquid is the main reason electrolyte-gated at −25 V for 1 h were studied by HAADF- for the poor cyclability of uncapped Co3O4 films. However, STEM and EELS. As shown in Figure 5(a)−(c), an when a Ta interlayer is grown adjacent to Co3O4, O2− ions inhomogeneous microstructure inside the Co3O4 layer is move back and forth along relatively short distances (for generated upon voltage application, with Co-rich and O-rich alternating applied voltages of opposite polarities), and the regions, leading to void-like morphologies, in agreement with process is confined between Co3O4 and the TaOx solid previous studies on this kind of magneto-ionic material.11 In electrolyte, thus drastically improving cyclability. turn, an enrichment in O is observed in the upper part of the To further understand the effect of the top Ta layer on the Ta layer. This is clearly evidenced by quantitative EELS induced magnetic properties from the perspective of the ion analysis (Figure 5(d)) and high-resolution TEM (Figure 5(e)). 6979 https://doi.org/10.1021/acsnano.3c01105 ACS Nano 2023, 17, 6973−6984 ACS Nano www.acsnano.org Article Figure 7. Positron annihilation spectroscopy characterization of the as-grown and voltage-treated (−25 V for 1 h) Co3O4/Ta (25 nm) sample. (a) Dependence of the low electron momentum fraction (S) on the positron implantation energy (Ep). (b) Dependence of the average positron lifetime, τav, on Ep. (c) Dependence of the positron lifetime components (τi) on Ep. (d) Dependence of the relative defect type intensity (Ii) on Ep.. This indicates that, besides the O2− ion transport from Co3O4 consequent oxidation of Cu to Cu2O. The CV curve of Co3O4 toward and across the Ta layer, O2− ions also locally migrate grown onto the Ti/Cu does not show the Cu waves, and redistribute inside the Co3O4 layer, eventually forming confirming that the cobalt oxide layer does cover completely metallic Co.11 Interestingly, the mixing of O and Ta elements the Cu phase and prevents its oxidation/reduction. Once Ta is on the top part of the Ta layer demonstrates the oxidation of grown on Co3O4, a tiny wide oxidation wave is observed, Ta. Considering that the liquid electrolyte used in this work centered at 0.6 V vs Pt, in agreement with the expected provides a nonaqueous environment, the observed O signal oxidation of Ta. Since both coatings (Co3O4 and TaOx) are can only originate from oxygen transport from Co3O4. The less conductive than Cu, the redox processes are hindered accumulation of O2− on the top of the TaOx layer suggests that compared to the case of Cu. ion transport remains rather restricted within the Co3O4/TaOx To shed further light on the microstructure of the films upon bilayers (i.e., short-distance ion diffusion). In other words, the magneto-ionic actuation, Doppler broadening variable energy presence of the thin TaOx solid electrolyte limits the amount positron annihilation spectroscopy (DB-VEPAS)55 and varia- of O2− released to the liquid, thereby enhancing cyclability. ble energy positron annihilation lifetime spectroscopy Furthermore, during this magneto-ionic experiment, Co3O4 (VEPALS)56 experiments were performed (see Experimental tends to become more amorphous. This is corroborated by θ/ Section). PAS is a sensitive probe to open volume defects on 2θ X-ray diffraction (see Figure 6(a)), where the peak from atomic scales and across depth due to positrons’ preferential Co3O4 (111) planes disappears upon voltage treatment. In localization and subsequent annihilation with electrons in addition, the FFT spots obtained from high-resolution TEM crystal empty spaces, i.e., vacancies and their agglomerates. The prove the existence of metallic Co in the treated sample, as can energy distribution and time necessary for annihilation be seen in Figure 6(b). The formation of metallic Co is depends on the local electron density.57 As shown in Figure responsible for the MS increase after voltage treatment (Figure 7(a), after negative biasing at −25 V, the low electron 2(c) and Figure 3). momentum fraction, S (directly proportional to the size and To assess the electrochemical behavior of the system, cyclic concentration of defects),58 increases in the Ta region but voltammetry (CV) curves were recorded for all the surfaces slightly drops in the Co3O4 layer, suggesting a substantial raise involved in the main magneto-ionic setup (see Figure S1 in the of defect density in Ta and a slight drop of defect Supporting Information). When the Cu grown on Ti is concentration in Co3O4. The increase of defect density in Ta exposed to the electrolyte, a cathodic wide peak is observed, is probably a consequence of the O2− interdiffusion and the centered at −0.9 V vs Pt reference electrode, while further formation of TaOx. Additionally, the average defect size (which anodic runs evidence a flat peak. This is consistent, in is proportional to the average positron lifetime, τav), strongly principle, with the reduction of Cu2O (native oxide layer increases in both TaOx and Co3O4 layers after voltage formed on Cu when exposed to air) to Cu (first wave) and actuation (Figure 7(b)). The expected average size is in the 6980 https://doi.org/10.1021/acsnano.3c01105 ACS Nano 2023, 17, 6973−6984 ACS Nano www.acsnano.org Article range of large vacancy agglomerates (8−10 vacancies).59 being released to the liquid electrolyte, i.e., limiting oxygen ion Figure 7(b) and (c) show the existence of three discrete transport within the Co3O4/TaOx layers. This is confirmed by lifetime components (τ1, τ2, and τ3),59 which correspond to compositional/structural characterization using HAADF- three different average defect sizes.11 They were obtained by STEM and EELS as well as positron spectroscopy experiments. deconvolution of PALS spectra using a nonlinear least-squared Beyond magneto-ionics, the reported strategy to enhance fitting method (package PALSfit software).60 The correspond- cyclability can be easily extrapolated to other systems relying ing relative intensities (I1, I2, and I3) relate to the concentration on ion transport mechanisms, such as iontronics, sensors, or of each defect type (size). The larger τi is, the larger the defect neuromorphic computing. size is since it takes longer for positrons to be annihilated with electrons.57 In the as-grown sample, only τ1 and τ2 lifetime EXPERIMENTAL SECTION components are detected, meaning an absence of large void- Sample Fabrication. Co3O4 thin films of 40 nm thickness were like structures (no τ3) in the prebiasing state. The most grown at room temperature by reactive sputtering in a high-vacuum abundant defect size (τ1) contributes as I1 = 62−77% of the chamber (with a base pressure of <8 × 10−8 Torr) on nondoped overall signal and is in the range of bivacancy clusters in the ⟨100⟩-oriented Si wafers previously coated with a 20-nm-thick upper subsurface Ta region (i.e., passivation layer) and single titanium adhesion layer and 60-nm-thick copper seed layer. Prior to vacancy in the underneath Ta sublayer.59 τ1 ≈ 213 ps for growing Co3O4, the Cu seed layers were partly masked to leave enough space for the electric contact (i.e., to later serve as a working Co3O4 represents a defect size involving 3−4 mixed (Co and electrode). The Co3O4 films were subsequently coated with Ta O) vacancies within a complex, based on our previous works/ protective layers of variable thickness, ranging from 5 to 50 nm, which publications.11 After biasing at −25 V, τ1 strongly increases were left unprotected to self-passivate in air. Two reference films were both in Ta and Co3O4 layers. Open volume is generated in the also prepared to serve as references: uncoated Co3O4, and Co3O4 size range of 4−6 vacancy agglomerations for Ta and >7−8 for coated with Pt with the same range of thicknesses as Ta. The growth Co3O4, which results in an increase of intensity to I1 = 75− of Co3O4 was carried out in a mixed Ar and O2 atmosphere using an 95%. The second lifetime component, τ2, for the as-grown oxygen partial pressure of 50% and a total working pressure of 3 × sample is in the range of large vacancy agglomerations (>10 10−3 Torr. The distance between the substrate and targets was around vacancies), typical for grain boundaries, with corresponding I2 10 cm, and the growth rate was approximately 0.6 Å s−1. Ti, Cu, Ta, and Pt layers were grown under 3 × 10−3 Torr Ar. The gun power to = 28−37% of positrons being annihilated at these defect states. grow Ta was 100 W. Negative biasing increases τ2 to the range of voids (the Magneto-Ionic Characterization and Cyclic Voltammetry threshold is usually about 500 ps) with an average diameter in Curves. Room-temperature magnetoelectric measurements were the range of 0.28−0.37 nm,61 while their density is quite small, performed using a commercial vibrating sample magnetometer from up to I2 ≈ 10%, and they do not reach deeper than the upper Micro Sense (LOT, Quantum Design), with a maximum applied in- 10 nm of Ta. Clearly, these defects are associated with the plane magnetic field of 2 T. The samples were electrolyte-gated using formation of the upper TaOx (Figure 5). Finally, a larger pore an external Agilent B2902A power supply, applying voltage between population (0.7−0.8 nm in diameter) was found after biasing the counter electrode (a Pt wire) and the working electrode (e.g., the investigated Si/Ti/Cu/Co3O4/Ta thin films) in a homemade (τ3), with low but not negligible intensity (I3 ≈ 1.4%), which electrolytic cell (see Figure 2(b)). The electrolyte consisted of extends into the Co3O4 layer. The existence of pore-related anhydrous propylene carbonate with Na+ and OH− solvated species components (especially τ3) is a fingerprint of increasing (10−25 ppm), formed by immersing small pieces of metallic sodium amorphization of the TaOx layer, whereas the Co3O4 layer that were able to react with any possible traces of water.11 Negative remains less affected and nanocrystalline. The largest free voltages in this work indicate the accumulation of negative charges at volume is found in the direct vicinity of the surface, i.e., the the working electrode (and vice versa for positive voltages). The most amorphized region. The increase of positron lifetimes in magnetization (M) was obtained by normalizing the magnetic the Ta layer after biasing evidences the O2− transport- moment to the sample volume exposed to the electrolyte. Note that generated expansion of available open volume channels, the linear slopes in the hysteresis loops at high fields (arising from diamagnetic or paramagnetic contributions) were subtracted by which enables large cyclability of the system. correcting the background signal (i.e., at fields always significantly larger than the saturation fields). CONCLUSIONS Cyclic voltammetry curves for all the surfaces involved in the main In summary, this work demonstrates the beneficial effect of magneto-ionic setup were also recorded. For these experiments, a scan adding a thin solid ionic conductor (in this case, a naturally speed of 10 mV/s and a potential sweep range from 0 V to −2 V to passivated amorphous Ta interlayer) on the magneto-ionic +1.4 V and to 0 V vs Pt were selected. The exposed surface was 0.5 response of Co3O4 films under the action of an electric field cm2, and the distance between the reference and working electrode was 4 mm, thus mimicking the conditions utilized during the when immersed in liquid electrolytes. A priori, adding a magneto-ionic experiments. Here, however, we used three electrodes capping layer between the Co3O4 film and the liquid (including a reference electrode), whereas magneto-ionic experiments electrolyte could be thought of as simply hindering magneto- were performed in a two-electrode configuration. Thus, all voltages in ionics and reducing (or eventually suppressing) any oxygen ion the CV curves are given with respect to a reference electrode, in this migration triggered by the externally applied electrical voltage. case Pt (99.99% Goodfellow), known to act well in organic solvents. This is indeed observed for thin metallic capping layers (e.g., 5 The global cell (two-electrode) potential has been measured during nm Pt) or thick highly resistive Ta layers (with a thickness these CV experiments to be on the order of 4 to 6 V depending on the larger than 50 nm). However, for thinner Ta interlayers, in system. Further potential could not be applied during the CV curves spite of a moderate reduction of the steady-state MS (by a due to the limitations of the utilized potentiostat (VSP Biologic). However, the curves obtained in this potential range are already factor of 25% for 25-nm-thick Ta), a drastic enhancement of representative of the electrochemical behavior of the system. magneto-ionic cyclability is observed, from less than 30 cycles Structural and Compositional Measurements. θ/2θ XRD in uncoated Co3O4 to more than 800 cycles for a Ta thickness patterns were collected on a Materials Research diffractometer from of 25 nm. Such enhancement of endurance is ascribed to the Malvern PANalytical Company, equipped with a PIXcel1D detector, key role of the generated TaOx layer in preventing O2− from using Cu Kα radiation. HRTEM, HAADF-STEM, and EELS were 6981 https://doi.org/10.1021/acsnano.3c01105 ACS Nano 2023, 17, 6973−6984 ACS Nano www.acsnano.org Article carried out on a TECNAI F20 HRTEM/STEM microscope operated AUTHOR INFORMATION at 200 kV. Cross-sectional lamellae were prepared by focused ion Corresponding Authors beam, placed onto a copper transmission electron microscopy grid, and topped with a protective platinum layer. Enric Menéndez − Departament de Física, Universitat Doppler Broadening Variable Energy Positron Annihilation Autònoma de Barcelona, 08193 Cerdanyola del Vallès, Spectroscopy and Variable Energy Positron Annihilation Spain; orcid.org/0000-0003-3809-2863; Lifetime Spectroscopy. DB-VEPAS measurements were conducted Email: enric.menendez@uab.cat at the setup for in situ defect analysis (AIDA)55 of the slow positron Jordi Sort − Departament de Física, Universitat Autònoma de beamline (SPONSOR).62 Positrons were accelerated and mono- Barcelona, 08193 Cerdanyola del Vallès, Spain; Institució energetically implanted into samples in the range of Ep = 0.05−35 Catalana de Recerca i Estudis Avançats (ICREA), E-08010 keV, which allows for depth-sensitive analysis. The mean positron Barcelona, Spain; orcid.org/0000-0003-1213-3639; implantation depth was approximated using a simple material density Email: jordi.sort@uab.cat (ρ)-dependent formula: ⟨z⟩ = 36/ρ·Ep1.62.63 Since at the annihilation site thermalized positrons have very small momentum compared to Authors the electrons, a broadening of the 511 keV line is observed mostly due Zhengwei Tan − Departament de Física, Universitat to momentum of the electrons, which is measured with a high-purity Autònoma de Barcelona, 08193 Cerdanyola del Vallès, Spain Ge detector (overall energy resolution of 1.09 ± 0.01 at 511 keV). Zheng Ma − Departament de Física, Universitat Autònoma de This broadening is characterized by a parameter S defined as a Barcelona, 08193 Cerdanyola del Vallès, Spain fraction of the annihilation distribution in the middle (511 ± 0.93 Laura Fuentes − Institut de Ciència de Materials de Barcelona, keV). The S-parameter is a fraction of positrons annihilating with low- CSIC, 08193 Bellaterra, Barcelona, Spain; Centre Nacional momentum valence electrons and represents vacancy-type defects and de Microelectrònica, Institut de Microelectrònica de their concentration.58 VEPALS measurements were conducted at the Barcelona-CSIC, 08193 Bellaterra, Barcelona, Spain; monoenergetic positron source (MePS) beamline, which is an end orcid.org/0000-0002-8799-2369 station of the radiation source ELBE (Electron Linac for beams with Maciej Oskar Liedke − Institute of Radiation Physics, high Brilliance and low Emittance) at Helmholtz-Zentrum Dresden- Helmholtz-Zentrum Dresden - Rossendorf, Dresden 01328, Rossendorf (Germany).56 A digital lifetime CeBr3 scintillator detector was used, with a homemade software employing an SPDevices Germany; orcid.org/0000-0001-7933-7295 ADQ14DC-2X with 14-bit vertical resolution and 2 GS s −1 Maik Butterling − Institute of Radiation Physics, Helmholtz- (gigasamples per second) horizontal resolution and with a time Zentrum Dresden - Rossendorf, Dresden 01328, Germany resolution function down to about 0.230 ns.64 The resolution function Ahmed G. Attallah − Institute of Radiation Physics, required for spectrum analysis uses two Gaussian functions with Helmholtz-Zentrum Dresden - Rossendorf, Dresden 01328, distinct intensities depending on the positron implantation energy, Ep, Germany; orcid.org/0000-0002-7759-0315 and appropriate relative shifts. All spectra contained at least 1 × 107 Eric Hirschmann − Institute of Radiation Physics, Helmholtz- counts. The spectra were deconvoluted using the PALSfit fitting Zentrum Dresden - Rossendorf, Dresden 01328, Germany software into discrete lifetime components, which directly confirm Andreas Wagner − Institute of Radiation Physics, Helmholtz- different defect types (i.e., sizes).60 Typical lifetime spectrum N(t) is Zentrum Dresden - Rossendorf, Dresden 01328, Germany described by N(t) = ∑(1/τi)Ii exp(−t/τi), where τi and Ii are the Llibertat Abad − Centre Nacional de Microelectrònica, Institut positron lifetime and intensity of the ith component, respectively (∑Ii de Microelectrònica de Barcelona-CSIC, 08193 Bellaterra, = 1). Barcelona, Spain The corresponding relative intensities (Ii) reflect to a large extent Nieves Casañ-Pastor − Institut de Ciència de Materials de the concentration of each defect type, and positron lifetimes (τi) are Barcelona, CSIC, 08193 Bellaterra, Barcelona, Spain; directly proportional to defect size (i.e., the larger the open volume, orcid.org/0000-0003-2979-4572 the lower the probability and the longer it takes for positrons to be Aitor F. Lopeandia − Departament de Física, Universitat annihilated with electrons). The positron lifetime and its intensity Autònoma de Barcelona, 08193 Cerdanyola del Vallès, were probed as a function of positron implantation energy Ep or, in Spain; Catalan Institute of Nanoscience and Nanotechnology other words, implantation depth (thickness). The average positron (ICN2), CSIC and BIST, 08193 Barcelona, Spain lifetime τav is defined as τav = ∑iτi·Ii, which is the weighted average of the defect size. The shortest lifetime component (τ1 < 0.32 ns) Complete contact information is available at: represents positron annihilation inside vacancy clusters (likely within https://pubs.acs.org/10.1021/acsnano.3c01105 grains) and/or at the grain boundaries, depending on the film’s microstructure. The intermediate lifetime (0.35 < τ2 < 0.90 ns) Author Contributions accounts for annihilation at larger vacancy clusters (linked to grain The manuscript was written through contributions of all boundaries and their intersections), surface states, and small voids/ authors. All authors have given approval to the final version of pores (0.28−0.37 nm in diameter, calculated based on the shape-free the manuscript. model for pore-size estimation of Wada et al.;59 the longest lifetime Notes component (2.3 < τ3 < 3.3 ns) indicates contributions of larger voids (0.58−0.74 nm in diameter). The authors declare no competing financial interest. ACKNOWLEDGMENTS ASSOCIATED CONTENT Financial support by the European Union’s Horizon 2020 * sı Supporting Information Research and Innovation Programme (“BeMAGIC” European The Supporting Information is available free of charge at Training Network, ETN/ITN Marie Skłodowska−Curie Grant https://pubs.acs.org/doi/10.1021/acsnano.3c01105. No. 861145), the European Research Council (2021-ERC- Advanced “REMINDS” Grant No. 101054687), the Spanish Cyclic voltammetry curves of the basic investigated layer Government (CEX2019-000917-S y PID2021-123276OB-I00, structures (PDF) PID2020-116844RB-C21, and PDC2021-121276-C31), and 6982 https://doi.org/10.1021/acsnano.3c01105 ACS Nano 2023, 17, 6973−6984 ACS Nano www.acsnano.org Article the Generalitat de Catalunya (2021-SGR-00651) is acknowl- Ruddlesde-Popper-Type Manganite by Electrochemical Fluoride Ion edged. J.S. thanks the Spanish “Fábrica Nacional de Moneda y Intercalation. Adv. Electron. Mater. 2020, 6, 1900974. Timbre” (FNMT) for fruitful discussions. E.M. is a Serra (18) Tan, A. J.; Huang, M.; Avci, C. O.; Büttner, F.; Mann, M.; Hu, Húnter Fellow. Parts of this research were carried out at ELBE W.; Mazzoli, C.; Wilkins, S.; Tuller, H. L.; Beach, G. S. D. Magneto- Ionic Control of Magnetism Using a Solid-State Proton Pump. Nat. at the Helmholtz-Zentrum Dresden - Rossendorf e. V., a Mater. 2019, 18, 35−41. member of the Helmholtz Association. We would like to thank (19) Ye, X.; Singh, H. K.; Zhang, H.; Geßwein, H.; Chellali, M. R.; the facility staff for assistance. This work was partially Witte, R.; Molinari, A.; Skokov, K.; Gutfleisch, O.; Hahn, H.; Kruk, R. supported by the Impulse-und Net-working fund of the Giant Voltage-Induced Modification of Magnetism in Micron-Scale Helmholtz Association (FKZ VH-VI-442 Memriox) and the Ferromagnetic Metals by Hydrogen Charging. Nat. Commun. 2020, Helmholtz Energy Materials Characterization Platform 11, 4849. (03ET7015). (20) Chen, G.; Ophus, C.; Quintana, A.; Kwon, H.; Won, C.; Ding, H.; Wu, Y.; Schmid, A. K.; Liu, K. Reversible Writing/deleting of Magnetic Skyrmions Through Hydrogen Adsorption/desorption. Nat. REFERENCES Commun. 2022, 13, 1350. (1) Shalf, J. M.; Leland, R. Computing Beyond Moore’s Law. (21) Lu, N.; Zhang, Z.; Wang, Y.; Li, H.-B.; Qiao, S.; Zhao, B.; He, Computer 2015, 48, 14−23. Q.; Lu, S.; Li, C.; Wu, Y.; Zhu, M.; Lyu, X.; Chen, X.; Li, Z.; Wang, (2) Ramesh, R.; Manipatruni, S. Electric Field Control of M.; Zhang, J.; Tsang, S. C.; Guo, J.; Yang, S.; Zhang, J.; Deng, K.; Magnetism. Proc. R. Soc. A 2021, 477, 20200942. Zhang, D.; Ma, J.; Ren, J.; Wu, Y.; Zhu, J.; Zhou, S.; Tokura, Y.; Nan, (3) Ž utić, I.; Fabian, J.; Das Sarma, S. Spintronics: Fundamentals C.-W.; Wu, J.; Yu, P. Enhanced Low-Temperature Proton and Applications. Rev. Mod. Phys. 2004, 76, 323−410. Conductivity in Hydrogen-Intercalated Brownmillerite Oxide. Nat. (4) Wolf, S. A.; Awschalom, D. D.; Buhrman, R. A.; Daughton, J. M.; Energy 2022, 7, 1208−1216. von Molnar, S.; Roukes, M. L.; Chtchelkanova, A. Y.; Treger, D. M. (22) De Rojas, J.; Quintana, A.; Lopeandía, A.; Salguero, J.; Muñiz, Spintronics: A Spin-Based Electronics Vision for the Future. Science B.; Ibrahim, F.; Chshiev, M.; Nicolenco, A.; Liedke, M. O.; Butterling, 2001, 294, 1488−1495. M.; Wagner, A.; Sireus, V.; Abad, L.; Jensen, C. J.; Liu, K.; Nogués, J.; (5) Kawahara, T.; Ito, K.; Takemura, R.; Ohno, H. Spin-Transfer Costa-Krämer, J. L.; Menéndez, E.; Sort, J. Voltage-Driven Motion of Torque RAM Technology: Review and Prospect. Microelectron. Reliab. Nitrogen Ions: A New Paradigm for Magneto-Ionics. Nat. Commun. 2012, 52, 613−627. 2020, 11, 5871. (6) Brataas, A.; Kent, A. D.; Ohno, H. Current-Induced Torques in (23) De Rojas, J.; Salguero, J.; Ibrahim, F.; Chshiev, M.; Quintana, Magnetic Materials. Nat. Mater. 2012, 11, 372−381. A.; Lopeandia, A.; Liedke, M. O.; Butterling, M.; Hirschmann, E.; (7) Han, X.; Wang, X.; Wan, C.; Yu, G.; Lv, X. Spin-Orbit Torques: Wagner, A.; Abad, L.; Costa-Krämer, J. L.; Menéndez, E.; Sort, J. Materials, Physics, and Devices. Appl. Phys. Lett. 2021, 118, 120502. Magneto-ionics in Single-layer Transition Metal Nitrides. ACS Appl. (8) Gambardella, P.; Miron, I. M. Current-Induced Spin-orbit Mater. Interfaces 2021, 13, 30826−30834. Torques. Philos. Trans. R. Soc. A 2011, 369, 3175−3197. (24) De Rojas, J.; Salguero, J.; Quintana, A.; Lopeandia, A.; Liedke, (9) Shiota, Y.; Nozaki, T.; Bonell, F.; Murakami, S.; Shinjo, T.; M. O.; Butterling, M.; Attallah, A. G.; Hirschman, E.; Wagner, A.; Suzuki, Y. Induction of Coherent Magnetization Switching in a Few Abad, L.; Costa-Krämer, J. L.; Sort, J.; Menéndez, E. Critical Role of Atomic Layers of FeCo Using Voltage Pulses. Nat. Mater. 2012, 11, Electrical Resistivity in Magnetoionics. Phys. Rev. Appl. 2021, 16, 39−43. 034042. (10) Song, C.; Cui, B.; Li, F.; Zhou, X.; Pan, F. Recent Progress in (25) Tan, Z.; Martins, S.; Escobar, M.; De Rojas, J.; Ibrahim, F.; Voltage Control of Magnetism: Materials, Mechanisms, and Perform- Chshiev, M.; Quintana, A.; Lopeandia, A.; Costa-Krämer, J. L.; ance. Prog. Mater. Sci. 2017, 87, 33−82. Menéndez, E. From Binary to Ternary Transition-metal Nitrides: A (11) Quintana, A.; Menéndez, E.; Liedke, M. O.; Butterling, M.; Boost Toward Nitrogen Magneto-ionics. ACS Appl. Mater. Interfaces Wagner, A.; Sireus, V.; Torruella, P.; Estradé, S.; Peiró, F.; 2022, 14, 44581−44590. Dendooven, J.; Detavernier, C.; Murray, P. D.; Gilbert, D. A.; Liu, (26) Tan, Z.; De Rojas, J.; Martins, S.; Lopeandia, A.; Quintana, A.; K.; Pellicer, E.; Nogués, J.; Sort, J. Voltage-Controlled ON−OFF Cialone, M.; Herrero-Martín, J.; Meersschaut, J.; Vantomme, A.; Ferromagnetism at Room Temperature in a Single Metal Oxide Film. Costa-Krämer, J. L.; Sort, J.; Menéndez, E. Frequency-dependent ACS Nano 2018, 12, 10291−10300. Stimulated and Post-stimulated Voltage Control of Magnetism in (12) Bauer, U.; Yao, L.; Tan, A. J.; Agrawal, P.; Emori, S.; Tuller, H. Transition Metal Nitrides: Towards Brain-Inspired Magneto-ionics. L.; Van Dijken, S.; Beach, G. S. D. Magneto-ionic Control of Mater. Horiz. 2023, 10, 88−96. Interfacial Magnetism. Nat. Mater. 2015, 14, 174−181. (27) Navarro-Senent, C.; Quintana, A.; Menéndez, E.; Pellicer, E.; (13) Gilbert, D. A.; Grutter, A. J.; Arenholz, E.; Liu, K.; Kirby, B. J.; Sort, J. Electrolyte-gated Magnetoelectric Actuation: Phenomenology, Borchers, J. A.; Maranville, B. B. Structural and Magnetic Depth Materials, Mechanisms, and Prospective Applications. APL Mater. Profiles of Magneto-ionic Heterostructures Beyond the Interface 2019, 7, 030701. Limit. Nat. Commun. 2016, 7, 12264. (28) De Rojas, J.; Quintana, A.; Rius, G.; Stefani, C.; Domingo, N.; (14) Rojas, J.; Quintana, A.; Lopeandia, A.; Salguero, J.; Costa- Costa-Krämer, J. L.; Menéndez, E.; Sort, J. Voltage Control of Krämer, J. L.; Abad, L.; Liedke, M. O.; Butterling, M.; Wagner, A.; Magnetism with Magneto-Ionic Approaches: Beyond Voltage-Driven Henderick, L.; Dendooven, J.; Detavernier, C.; Sort, J.; Menéndez, E. Oxygen Ion Migration. Appl. Phys. Lett. 2022, 120, 070501. Boosting Room-Temperature Magneto-ionics in a Non-Magnetic (29) Zehner, J.; Wolf, D.; Hasan, M. U.; Huang, M.; Bono, D.; Oxide Semiconductor. Adv. Funct. Mater. 2020, 30, 2003704. Nielsch, K.; Leistner, K.; Beach, G. S. D. Magnetoionic Control of (15) Martins, S.; De Rojas, J.; Tan, Z.; Cialone, M.; Lopeandia, A.; Perpendicular Exchange Bias. Phys. Rev. Mater. 2021, 5, L061401. Herrero-Martín, J.; Costa-Krämer, J. L.; Menéndez, E.; Sort, J. (30) Martins, S.; Ma, Z.; Solans-Monfort, X.; Sodupe, M.; Dynamic Electric-field-induced Magnetic Effects in Cobalt Oxide Rodriguez-Santiago, L.; Menéndez, E.; Pellicer, E.; Sort, J. Enhancing Thin Films: Towards Magneto-ionic Synapses. Nanoscale 2022, 14, Magneto-ionic Effects in Cobalt Oxide Films by Electrolyte 842−852. Engineering. Nanoscale Horiz. 2022, 8, 118−126. (16) Dasgupta, S.; Das, B.; Knapp, M.; Brand, R. A.; Ehrenberg, H.; (31) Zhang, Q.; Luo, X.; Wang, L.; Zhang, L.; Khalid, B.; Gong, J.; Kruk, R.; Hahn, H. Intercalation-driven Reversible Control of Wu, H. Lithium-Ion Battery Cycling for Magnetism Control. Nano Magnetism in Bulk Ferromagnets. Adv. Mater. 2014, 26, 4639−4644. Lett. 2016, 16, 583−587. (17) Vasala, S.; Jakob, A.; Wissel, K.; Waidha, A. I.; Alff, L.; (32) Quintana, A.; Firme, A. A.; Jensen, C. J.; Zheng, D.; Liu, C.; Clemens, O. Reversible Tuning of Magnetization in a Ferromagnetic Zhang, X.; Liu, K. Hydroxide-Based Magneto-ionics: Electric-Field 6983 https://doi.org/10.1021/acsnano.3c01105 ACS Nano 2023, 17, 6973−6984 ACS Nano www.acsnano.org Article Control of Reversible Paramagnetic-to-Ferromagnetic Switch in α- (50) Krikorian, E.; Sneed, R. J. Deposition of Tantalum, Tantalum Co(OH)2 films. J. Mater. Chem. C 2022, 10, 17145. Oxide, and Tantalum Nitride with Controlled Electrical Character- (33) Walter, J.; Voigt, B.; Day-Roberts, E.; Heltemes, K.; Fernandes, istics. J. Appl. Phys. 1996, 37, 3674−3681. R. M.; Birol, T.; Leighton, C. Voltage-induced Ferromagnetism in a (51) Gerstenberg, D.; Calbick, C. J. Effects of Nitrogen, Methane, Diamagnet. Sci. Adv. 2020, 6, eabb7721. and Oxygen on Structure and Electrical Properties of Thin Tantalum (34) Molinari, A.; Hahn, H.; Kruk, R. Voltage-control of Magnetism Films. J. Appl. Phys. 1964, 35, 402−407. in All-solid-state and Solid/liquid Magnetoelectric Composites. Adv. (52) De Vries, J. W. C. Temperature and Thickness Dependence of Mater. 2019, 31, 1806662. the Resistivity of Thin Polycrystalline Aluminium, Cobalt, Nickel, (35) Pachat, R.; Ourdani, D.; Syskaki, M.; Lamperti, A.; Roy, S.; Palladium, Silver and Gold Films. Thin Solid Films 1988, 167, 25−32. Chen, S.; Pietro, A. D.; Largeau, L.; Juge, R.; Massouras, M.; Balan, (53) Zubler, E. G. The Gettering Properties of Tantalum. J. C.; Van Der Jagt, J. W.; Agnus, G.; Roussigné, Y.; Gabor, M.; Chérif, Electrochem. Soc. Jpn. 1963, 110, 1072. S. M.; Durin, G.; Ono, S.; Langer, J.; Querlioz, D.; Ravelosona, D.; (54) Bean, C. P.; Fisher, J. C.; Vermilyea, D. A. Ionic Conductivity Belmeguenai, M.; Herrera Diez, L. Magneto-Ionics in Annealed W/ of Tantalum Oxide at Very High Fields. Phys. Rev. 1956, 101, 551. CoFeB/HfO2 Thin Films. Adv. Mater. Interfaces 2022, 9, 2200690. (55) Liedke, M. O.; Anwand, W.; Bali, R.; Cornelius, S.; Butterling, (36) Zhou, X.; Yan, Y.; Jiang, M.; Cui, B.; Pan, F.; Song, C. Role of M.; Trinh, T. T.; Wagner, A.; Salamon, S.; Walecki, D.; Smekhova, A. Oxygen Ion Migration in the Electrical Control of Magnetism in Pt/ Open Volume Defects and Magnetic Phase Transition in Fe60Al40 Co/Ni/HfO2 Films. J. Phys. Chem. C 2016, 120, 1633−1639. Transition Metal Aluminide. J. Appl. Phys. 2015, 117, 163908. (37) Herrera Diez, L.; Kruk, R.; Leistner, K.; Sort, J. Magnetoelectric (56) Wagner, A.; Butterling, M.; Liedke, M. O.; Potzger, K.; Krause- Materials, Phenomena, and Devices. APL Mater. 2021, 9, 050401. Rehberg, R. Positron Annihilation Lifetime and Doppler Broadening (38) Chen, G.; Mascaraque, A.; Jia, H.; Zimmermann, B.; Robertson, Spectroscopy at the ELBE Facility. AIP Conf. Proc. 2018, 1970, M.; Lo Conte, R.; Hoffmann, M.; González Barrio, M. A.; Ding, H.; 040003. (57) Tuomisto, F.; Makkonen, I. Defect Identification in Semi- Wiesendanger, R.; Michel, E. G.; Blügel, S.; Schmid, A. K.; Liu, K. conductors with Positron Annihilation: Experiment and Theory. Rev. Large Dzyaloshinskii-Moriya Interaction Induced by Chemisorbed Mod. Phys. 2013, 85, 1583−1631. Oxygen on a Ferromagnet Surface. Sci. Adv. 2020, 6, eaba4924. (58) Clement, M.; De Nijs, J. M. M.; Balk, P.; Schut, H.; Van Veen, (39) Chen, G.; Robertson, M.; Hoffmann, M.; Ophus, C.; Fernandes A. Analysis of Positron Beam Data by the Combined Use of the Cauduro, A. L.; Lo Conte, R.; Ding, H.; Wiesendanger, R.; Blügel, S.; Shape-and Wing-Parameters. J. Appl. Phys. 1996, 79, 9029−9036. Schmid, A. K.; Liu, K. Observation of Hydrogen-Induced (59) Xu, Q.; Popov, E.; Troev, T.; Zhang, J.; Dai, Y. Positron Dzyaloshinskii-Moriya Interaction and Reversible Switching of Lifetime Calculation of Vacancy Clusters in Tantalum Containing Magnetic Chirality. Phys. Rev. X 2021, 11, 021015. Hydrogen and Helium. J. Nucl. Mater. 2018, 506, 71−75. (40) Huang, M.; Usama Hasan, M.; Klyukin, K.; Zhang, D.; Lyu, D.; (60) Olsen, J. V.; Kirkegaard, P.; Pedersen, N. J.; Eldrup, M. Gargiani, P.; Valvidares, M.; Sheffels, S.; Churikova, A.; Büttner, F.; PALSfit: A New Program for the Evaluation of Positron Lifetime Zehner, J.; Caretta, L.; Lee, K.-Y.; Chang, J.; Wang, J.-P.; Leistner, K.; Spectra. Phys. Status Solidi 2007, 4, 4004−4006. Yildiz, B.; Beach, G. S. D. Voltage Control of Ferrimagnetic Order and (61) Wada, K.; Hyodo, T. A Simple Shape-Free Model for Pore-Size Voltage-assisted Writing of Ferrimagnetic Spin Textures. Nat. Estimation with Positron Annihilation Lifetime Spectroscopy. J. Phys. Nanotechnol. 2021, 16, 981−988. Conf. Ser. 2013, 443, 012003. (41) Zehner, J.; Huhnstock, R.; Oswald, S.; Wolff, U.; Soldatov, I.; (62) Anwand, W.; Brauer, G.; Butterling, M.; Kissener, H. R.; Ehresmann, A.; Nielsch, K.; Holzinger, D.; Leistner, K. Nonvolatile Wagner, A. Design and Construction of a Slow Positron Beam for Electric Control of Exchange Bias by a Redox Transformation of the Solid and Surface Investigations. Defect Diffus. Forum. 2012, 331, 25− Ferromagnetic Layer. Adv. Electron. Mater. 2019, 5, 1900296. 40. (42) Lecuyer, S.; Quemerais, A.; Jezequel, G. Composition of (63) Dryzek, J.; Horodek, P. GEANT4 Simulation of Slow Positron Natural Oxide Films on Polycrystalline Tantalum Using XPS Electron Beam Implantation Profiles. Nucl. Instruments Methods Phys. Res. Sect. Take-Off Angle Experiments. Surf. Interface Anal. 1992, 18, 257−261. B Beam Interact. with Mater. Atoms. 2008, 266, 4000−4009. (43) Levine, B. R.; Sporer, S.; Poggie, R. A.; Della Valle, C. J.; (64) Hirschmann, E.; Butterling, M.; Hernandez Acosta, U.; Liedke, Jacobs, J. J. Experimental and Clinical Performance of Porous M. O.; Attallah, A. G.; Petring, P.; Görler, M.; Krause-Rehberg, R.; Tantalum in Orthopedic Surgery. Biomaterials 2006, 27, 4671−4681. Wagner, A. A New System for Real-time Data Acquisition and Pulse (44) Stella, K.; Burstel, D.; Franzka, S.; Posth, O.; Diesing, D. Parameterization for Digital Positron Annihilation Lifetime Spec- Preparation and Properties of Thin Amorphous Tantalum Films trometers With High Repetition rates. J. Instrum. 2021, 16, 08001. Formed by Small e-beam Evaporators. J. Phys. D: Appl. Phys. 2009, 42, 135417. (45) Dalla Torre, J.; Gilmer, G. H.; Windt, D. L.; Kalyanaraman, R.; Baumann, F. H.; O’Sullivan, P. L.; Sapjeta, J.; DíAz De La Rubia, T.; Djafari Rouhani, M. Microstructure of Thin Tantalum Films Sputtered onto Inclined Substrates: Experiments and Atomistic Simulations. J. Appl. Phys. 2003, 94, 263−271. (46) Liu, Q.; Wu, F.; Mu, D.; Wu, B. A Theoretical Study on Na+ Solvation in Carbonate Ester and Ether Solvents for Sodium-ion Batteries. Phys. Chem. Chem. Phys. 2020, 22, 2164−2175. (47) Weisheit, M.; Fähler, S.; Marty, A.; Souche, Y.; Poinsignon, C.; Givord, D. Electric Field-Induced Modification of Magnetism in Thin-Film Ferromagnets. Science 2007, 315, 349−351. (48) Silva, C.; Vovk, A.; Da Silva, R. C.; Strichovanec, P.; Algarabel, P. A.; Gonçalves, A. P.; Borges, R. P.; Godinho, M.; Cruz, M. M. Magnetic Properties of Co−N Thin Films Deposited by Reactive Sputtering. Thin Solid Films 2014, 556, 125−127. (49) Li, Z. L.; Du, B. X.; Yang, Z. R.; Li, J. Effects of Crystal Morphology on Space Charge Transportation and Dissipation of Sic/ Silicone Rubber Composites. IEEE Trans. Dielectr. Electr. Insul. 2017, 24, 2616−2625. 6984 https://doi.org/10.1021/acsnano.3c01105 ACS Nano 2023, 17, 6973−6984