PhD - Leuven | Just now
Voltage-controlled exchange coupling for next-generation low-power MRAM
Growing demand for data storage and computing is pushing conventional memories toward their limits in terms of volatility and power consumption. This has accelerated the development of energy-efficient non-volatile memories. Spintronic technologies and particularly Magnetic Random Access Memory (MRAM) are emerging as promising candidates for next-generation memory.
The basic building block of MRAM is the Magnetic Tunnel Junction (MTJ), consisting of two ferromagnetic layers separated by an ultrathin oxide barrier. Information is stored in the magnetization of the free layer while the fixed layer serves as reference. Reading via Tunnel Magnetoresistance (TMR) effect where the MTJ resistance depends on the relative magnetization orientation of the two layers. Spin-Transfer Torque (STT)-MRAM is a mature technology that has been commercialized as a replacement for eFlash and SRAM [1]. Data write scheme operates via injecting spin-polarized current through the MTJ to switch the free-layer magnetization. Spin-Orbit Torque (SOT)-MRAM has emerged as a promising alternative where spin torque is generated by a current flowing through an adjacent heavy metal. By separating the write and read current paths, SOT-MRAM offers faster switching and improved reliability. However, both STT- and SOT-MRAM rely on current-induced switching, requiring large current densities that lead to significant write energy consumption. Therefore, reducing the switching energy while maintaining fast, reliable, and scalable operation remains a key challenge for future MRAM technologies.
In this context, voltage-controlled magnetization switching emerges as a key pathway toward ultra-low-power MRAM. Several approaches have been investigated to enable voltage-driven MRAM, leveraging electric-field effects to control magnetization switching [2]. Magnetoelectric switching offers the possibility of electric-field-driven magnetization reversal, but the integration of magnetoelectric materials with MTJs remains challenging. Switching based on ionic motion can induce large modulation of magnetic properties, but its relatively slow switching speed due to ion migration. Voltage-Controlled Magnetic Anisotropy (VCMA) is particularly attractive because it combines low-power operation, compatibility with MTJ technology. However, deterministic field-free switching remains difficult to achieve since VCMA mainly modulates magnetic anisotropy rather than directly generating a switching torque [2]. More recently, VCMA-driven magnetization switching has been demonstrated in a synthetic antiferromagnetic tri-layer. Here, the applied voltage changes the magnetic properties of one ferromagnetic layer, altering the balance between magnetic anisotropy and interlayer exchange coupling and thereby driving the transition between parallel and antiparallel states. Unlike conventional VCMA switching, this mechanism does not rely on precisely timed magnetization precession, enabling deterministic switching [3]. However, the concept has so far been demonstrated only in microscale devices without MTJ integration, leaving the scalability and practical application to low power MRAM largely unexplored.
At imec, we have recently demonstrated an electrical TMR-based approach to selectively detect the magnetization state of synthetic antiferromagnetic material. This proof-of-concept demonstrates the feasibility of electrically probing magnetic state of synthetic antiferromagnetic at the nanoscale [4]. As a PhD candidate, you will build on this foundation to investigate the physics of voltage-driven switching in synthetic antiferromagnetic structures, optimize device, material designs, and advance voltage-controlled magnetic memory toward a scalable technology for future memory applications.
This PhD project aims to investigate such mechanisms and develop novel device concepts for next-generation voltage-controlled MRAM. Imec utilizes a dedicated 300-mm wafer platform to develop voltage controlled-MRAM devices. Through advanced electrical and physical characterizations, this dissertation seeks to elucidate the underlying physics governing the role of exchange coupling in voltage-induced magnetization switching dynamics of scaled MTJs. The experimental study will be complemented by performing micromagnetic modelling activities using tools such as OOMMF and Mumax. This research will leverage an extensive toolset for magnetic and electrical characterization, including techniques such as VSM, PMOKE, magneto-optical Kerr microscopy, and Hprobe electrical probing setup. Close collaboration with imec's device engineers specializing in magnetic devices will ensure synergy between device characteristic insights and practical implementation, addressing the technological challenges in voltage-controlled MRAM.
We are seeking a highly motivated candidate with a background in physics or engineering, a strong interest in experimental nanodevice research, and a passion for advancing next-generation memory and logic technologies, which are gaining significant momentum under initiatives such as the European Chips Act [3].
Required background: Physics, Engineering Science
Type of work: Literature study (10%), experimental work (70%), modelling (20%)
Supervisor: Clement Merckling (KU Leuven)
Co-supervisor: Clement Merckling (KU Leuven)
Daily advisor: Daniele Narducci
The reference code for this position is 2027-042. Mention this reference code in your application.