Periodic Reporting for period 1 - MagnOxy (Magneto-ionic devices based on oxygen-ion solid state intercalation)
Okres sprawozdawczy: 2024-01-01 do 2025-12-31
Artificial intelligence and data-driven technologies are transforming society, but their rapid growth is also driving a sharp increase in energy consumption. A major reason is that today’s computers are built around the “von Neumann” architecture, where information must constantly move between separate memory and processing units. This data movement is slow and energy-intensive and limits the sustainability of advanced computing. Developing new hardware concepts that reduce energy consumption while maintaining performance is therefore a key technological and societal challenge.
The MagnOxy project addressed this challenge by exploring new materials and device concepts for energy-efficient, non-volatile memory and in-memory computing. The core idea was to control magnetic and electronic properties in oxide thin films using voltage-driven oxygen ion motion. Instead of switching states with electric currents, which dissipate significant power, MagnOxy uses small electrical voltages to reversibly move oxygen ions in solid materials. This controlled motion modifies the atomic structure and, in turn, the functional properties of the material. Such magneto-ionic control offers a promising route to ultra-low-power devices that retain information without standby energy.
To achieve this vision, MagnOxy followed a device-oriented pathway combining solid-state ionics, electrochemistry, magnetism and micro- and nanofabrication. The project aimed to establish solid-state platforms capable of controlling oxygen content in oxide thin films and directly correlating these changes with magnetic and functional properties. It further aimed to accelerate ionic switching and expand device functionalities by engineering ionic transport at the nanoscale. Finally, it sought to demonstrate scalability and technological relevance by adopting fabrication processes and substrates compatible with industrial microelectronics, including large-area deposition and device miniaturization.
By creating and validating oxide-based platforms in which oxygen ions can be precisely inserted and removed under voltage, MagnOxy contributes to the development of future memory and computing technologies that are faster, more efficient and more sustainable. In the longer term, these concepts are relevant for European priorities on energy-efficient digital technologies, supporting a pathway toward greener artificial intelligence hardware and advanced computing systems with reduced environmental footprint.
The MagnOxy project addressed this challenge by exploring new materials and device concepts for energy-efficient, non-volatile memory and in-memory computing. The core idea was to control magnetic and electronic properties in oxide thin films using voltage-driven oxygen ion motion. Instead of switching states with electric currents, which dissipate significant power, MagnOxy uses small electrical voltages to reversibly move oxygen ions in solid materials. This controlled motion modifies the atomic structure and, in turn, the functional properties of the material. Such magneto-ionic control offers a promising route to ultra-low-power devices that retain information without standby energy.
To achieve this vision, MagnOxy followed a device-oriented pathway combining solid-state ionics, electrochemistry, magnetism and micro- and nanofabrication. The project aimed to establish solid-state platforms capable of controlling oxygen content in oxide thin films and directly correlating these changes with magnetic and functional properties. It further aimed to accelerate ionic switching and expand device functionalities by engineering ionic transport at the nanoscale. Finally, it sought to demonstrate scalability and technological relevance by adopting fabrication processes and substrates compatible with industrial microelectronics, including large-area deposition and device miniaturization.
By creating and validating oxide-based platforms in which oxygen ions can be precisely inserted and removed under voltage, MagnOxy contributes to the development of future memory and computing technologies that are faster, more efficient and more sustainable. In the longer term, these concepts are relevant for European priorities on energy-efficient digital technologies, supporting a pathway toward greener artificial intelligence hardware and advanced computing systems with reduced environmental footprint.
The MagnOxy project developed and experimentally validated solid-state platforms for voltage-controlled magneto-ionic devices based on oxygen-ion intercalation in oxide thin films. The work combined thin-film growth, electrochemical control, advanced magnetic characterization and microfabrication in an integrated research approach.
Mixed ionic–electronic conducting oxide thin films were fabricated by pulsed laser deposition on oxide-ion conducting substrates. A fully solid-state configuration was implemented, enabling reversible insertion and extraction of oxygen ions under small applied voltages in the range of approximately +0.1 V to −1.0 V. Two complementary methodologies were established to quantify oxygen stoichiometry variations. Electrochemical charge integration enabled quantitative determination of oxygen exchange, while iono-optic impedance spectroscopy, developed within the project, allowed dynamic characterization of oxygen insertion processes.
Using these approaches, reversible tuning of oxygen content was demonstrated in perovskite oxides over a wide compositional range. In SrFe₀.₅Co₀.₅O₃₋δ thin films, the oxygen content was reversibly varied from approximately 3−δ ≈ 2.7 to 3−δ ≈ 2.25 inducing clear magnetic phase transitions. Magnetic measurements confirmed voltage-driven modulation between ferromagnetic and paramagnetic states. In La₀.₅Sr₀.₅FeO₃₋δ, electrochemical control enabled switching between antiferromagnetic and paramagnetic behavior, further validated by synchrotron-based imaging of magnetic domains. These results demonstrate direct and reversible voltage control of magnetic states through oxygen-ion motion in solid materials.
To enhance ionic transport and explore advanced functionalities, vertically aligned nanocomposite architectures were fabricated. These structures integrate oxygen-conducting pathways directly within the active magnetic layer, reducing diffusion distances and facilitating faster ionic processes. Tip-enhanced Raman spectroscopy was implemented and optimized to probe nanoscale variations in structure and composition with spatial resolution in the tens of nanometres. This provided direct insight into local phase distribution and phase heterogeneity and established the physical basis for accelerated magneto-ionic switching.
In addition, the project explored multifunctional oxide heterostructures in which electrochemical oxygen control enables tuning of interfacial magnetic interactions. These findings extend magneto-ionics beyond simple modulation of magnetization amplitude and open routes toward reconfigurable spintronic and neuromorphic device concepts.
To increase technological relevance, industry-compatible radio-frequency sputtering was implemented for the deposition of dense oxide electrolyte thin films. Homogeneous films were successfully deposited on 4-inch silicon wafers. The thin films exhibited functional ionic conductivity of approximately 10⁻⁵ S·cm⁻¹ at 325 °C, confirming their suitability for device applications. Microfabrication processes based on photolithography and ion milling were developed to pattern oxide heterostructures at micrometre scale without degrading functional properties. This demonstrates compatibility of magneto-ionic oxide systems with standard microelectronic processing routes and establishes a realistic pathway toward device integration.
Together, these achievements establish a solid scientific and technological foundation for next-generation energy-efficient magneto-ionic memory and in-memory computing devices.
Mixed ionic–electronic conducting oxide thin films were fabricated by pulsed laser deposition on oxide-ion conducting substrates. A fully solid-state configuration was implemented, enabling reversible insertion and extraction of oxygen ions under small applied voltages in the range of approximately +0.1 V to −1.0 V. Two complementary methodologies were established to quantify oxygen stoichiometry variations. Electrochemical charge integration enabled quantitative determination of oxygen exchange, while iono-optic impedance spectroscopy, developed within the project, allowed dynamic characterization of oxygen insertion processes.
Using these approaches, reversible tuning of oxygen content was demonstrated in perovskite oxides over a wide compositional range. In SrFe₀.₅Co₀.₅O₃₋δ thin films, the oxygen content was reversibly varied from approximately 3−δ ≈ 2.7 to 3−δ ≈ 2.25 inducing clear magnetic phase transitions. Magnetic measurements confirmed voltage-driven modulation between ferromagnetic and paramagnetic states. In La₀.₅Sr₀.₅FeO₃₋δ, electrochemical control enabled switching between antiferromagnetic and paramagnetic behavior, further validated by synchrotron-based imaging of magnetic domains. These results demonstrate direct and reversible voltage control of magnetic states through oxygen-ion motion in solid materials.
To enhance ionic transport and explore advanced functionalities, vertically aligned nanocomposite architectures were fabricated. These structures integrate oxygen-conducting pathways directly within the active magnetic layer, reducing diffusion distances and facilitating faster ionic processes. Tip-enhanced Raman spectroscopy was implemented and optimized to probe nanoscale variations in structure and composition with spatial resolution in the tens of nanometres. This provided direct insight into local phase distribution and phase heterogeneity and established the physical basis for accelerated magneto-ionic switching.
In addition, the project explored multifunctional oxide heterostructures in which electrochemical oxygen control enables tuning of interfacial magnetic interactions. These findings extend magneto-ionics beyond simple modulation of magnetization amplitude and open routes toward reconfigurable spintronic and neuromorphic device concepts.
To increase technological relevance, industry-compatible radio-frequency sputtering was implemented for the deposition of dense oxide electrolyte thin films. Homogeneous films were successfully deposited on 4-inch silicon wafers. The thin films exhibited functional ionic conductivity of approximately 10⁻⁵ S·cm⁻¹ at 325 °C, confirming their suitability for device applications. Microfabrication processes based on photolithography and ion milling were developed to pattern oxide heterostructures at micrometre scale without degrading functional properties. This demonstrates compatibility of magneto-ionic oxide systems with standard microelectronic processing routes and establishes a realistic pathway toward device integration.
Together, these achievements establish a solid scientific and technological foundation for next-generation energy-efficient magneto-ionic memory and in-memory computing devices.
The MagnOxy project advances the state of the art in magneto-ionic materials and device concepts by demonstrating controlled and reversible voltage-driven oxygen-ion modulation of magnetic properties in solid-state oxide thin films. While previous studies often relied on liquid electrolytes or limited proof-of-concept experiments, MagnOxy implemented fully solid-state platforms compatible with microfabrication and industrial processing. This represents an important step toward practical device integration.
A key scientific advancement is the quantitative and reversible control of oxygen stoichiometry over a wide compositional range using small applied voltages. The project established new experimental methodologies to measure and monitor oxygen insertion dynamics, including iono-optic impedance spectroscopy and nanoscale tip-enhanced Raman characterization. These techniques provide deeper insight into ionic processes at nanometre length scales and enable a more precise understanding of how ionic motion governs magnetic phase transitions. Such methodological developments extend beyond the specific materials studied and can be applied broadly to other ionotronic and functional oxide systems.
The project also demonstrated that magneto-ionic control can be integrated into complex oxide heterostructures, opening the possibility of multifunctional devices in which magnetic interactions are tuned electrically. This extends magneto-ionics beyond simple modulation of magnetization amplitude and moves the field closer to reconfigurable spintronic and neuromorphic architectures.
Importantly, the project addressed scalability by implementing industry-compatible sputtering techniques and demonstrating homogeneous thin-film deposition on 4-inch silicon wafers. The compatibility with silicon substrates and standard microfabrication processes significantly lowers the technological barrier toward integration into existing semiconductor platforms. This positions magneto-ionic oxide devices as a realistic candidate for future energy-efficient memory and in-memory computing technologies.
The potential impacts of these results are both scientific and technological. Scientifically, the project contributes to a deeper understanding of voltage-controlled phase transitions in correlated oxides and establishes solid-state magneto-ionics as a viable mechanism for low-power device operation. Technologically, the demonstrated compatibility with scalable fabrication routes supports future translation toward spintronic memory, neuromorphic computing hardware and energy-efficient logic components.
To ensure further uptake and success, several steps are required. Continued research is needed to optimise switching speed, long-term cycling stability and device reliability. Further demonstration at device-array level will be necessary to validate performance in realistic computing architectures. Collaboration with semiconductor and spintronics industry partners will facilitate technology transfer and assessment of integration constraints. Support for intellectual property protection and access to scale-up facilities will also be important for commercialisation pathways.
Overall, MagnOxy establishes a scientifically validated and technologically promising foundation for voltage-controlled oxide devices and contributes to the development of sustainable hardware solutions aligned with Europe’s digital and green transition objectives.
A key scientific advancement is the quantitative and reversible control of oxygen stoichiometry over a wide compositional range using small applied voltages. The project established new experimental methodologies to measure and monitor oxygen insertion dynamics, including iono-optic impedance spectroscopy and nanoscale tip-enhanced Raman characterization. These techniques provide deeper insight into ionic processes at nanometre length scales and enable a more precise understanding of how ionic motion governs magnetic phase transitions. Such methodological developments extend beyond the specific materials studied and can be applied broadly to other ionotronic and functional oxide systems.
The project also demonstrated that magneto-ionic control can be integrated into complex oxide heterostructures, opening the possibility of multifunctional devices in which magnetic interactions are tuned electrically. This extends magneto-ionics beyond simple modulation of magnetization amplitude and moves the field closer to reconfigurable spintronic and neuromorphic architectures.
Importantly, the project addressed scalability by implementing industry-compatible sputtering techniques and demonstrating homogeneous thin-film deposition on 4-inch silicon wafers. The compatibility with silicon substrates and standard microfabrication processes significantly lowers the technological barrier toward integration into existing semiconductor platforms. This positions magneto-ionic oxide devices as a realistic candidate for future energy-efficient memory and in-memory computing technologies.
The potential impacts of these results are both scientific and technological. Scientifically, the project contributes to a deeper understanding of voltage-controlled phase transitions in correlated oxides and establishes solid-state magneto-ionics as a viable mechanism for low-power device operation. Technologically, the demonstrated compatibility with scalable fabrication routes supports future translation toward spintronic memory, neuromorphic computing hardware and energy-efficient logic components.
To ensure further uptake and success, several steps are required. Continued research is needed to optimise switching speed, long-term cycling stability and device reliability. Further demonstration at device-array level will be necessary to validate performance in realistic computing architectures. Collaboration with semiconductor and spintronics industry partners will facilitate technology transfer and assessment of integration constraints. Support for intellectual property protection and access to scale-up facilities will also be important for commercialisation pathways.
Overall, MagnOxy establishes a scientifically validated and technologically promising foundation for voltage-controlled oxide devices and contributes to the development of sustainable hardware solutions aligned with Europe’s digital and green transition objectives.