Multiscale and multi-physics modeling for advanced magnetic memories

Leuven - PhD
Meer dan twee weken geleden

Unlock the future of low energy memories for tomorrow's computers

With the event of the internet of things and of artificial intelligence, there is an important need to develop low energy efficient stand-alone and embedded memories. With that respect, memories based on the manipulation of the spin offer promising prospects. For instance, memories such as Spin-Transfer Torque Magnetic Random-Access Memory (STT-MRAM), where, the spin of the electrons is flipped using a spin-polarized current which is created by passing a current though a thin magnetic layer offers up to 60% of reduction of the energy consumption compared to a classical DRAM one. The effect is achieved thanks to a magnetic tunnel junction (MTJ) or a spin-valve, and STT-MRAM devices use STT tunnel junctions (STT-MTJ). This current is then directed into a thinner magnetic layer which transfers the angular momentum to the thin layer which changes its spin. To be able to achieve this spin manipulation, memories such as the STT-MRAM ones, are built from a complex stack of different materials (up to twenty different ones) with film thicknesses ranging from a sub-nanometer to 100 nm.

While the working principle of MRAM memories is a well-established process, the development of the next generation of magnetic memories is driven by the selection of the right materials and the quality of their interfaces. Improvements are hence bound to the understanding of the fundamental aspects that define material interfaces and to the establishment of device design rules that allow an improved efficient spin manipulation. Unfortunately, little is known on the factors that are driving the global spin response at the device level. The main problem is that there is a missing link between the macroscopic device reality, governed by continuum formalisms, and atomistic realities dictated by local interactions. A proper connection of these theories then needs to be built to describe the physics that governs the full magnetic coupling of multi-scaled materials.

Through this Ph.D. proposal, we aim fulfilling this gap by building links between atomistic simulations and micromagnetic formalisms used to model magnetic devices to study fundamental relations between material, interface magnetic properties and device performances. For instance, the magnetic properties of nano-confined materials will be studied using atomistic simulations (such as density functional theory), deriving information of the impact of confinement of metals on properties such as the exchange coupling and the magnetic anisotropy. The output generated will then be coupled with micromagnetic techniques to build the insights needed into the engineering of interfaces for the development of new memories.
IMEC will provide training to both UNIX/Linux and to the material modeling techniques. A strong motivation, a good knowledge of solid-state physics and/or magnetism and of UNIX/LINUX are a plus. Excellent writing and oral communication skills are desired.

Required background: Material engineering or physics with strong interest in numerical analysis

Type of work: literature 10 % and 90 % modeling

Supervisor: Michel Houssa

Daily advisor: Kiroubanand Sankaran

The reference code for this position is 2020-018. Mention this reference code on your application form.
Chinese nationals who wish to apply for the CSC scholarship, should use the following code when applying for this topic: CSC2020-11.


Share this on


Deze website maakt gebruik van cookies met als enige doel het analyseren van surfgedrag, zonder enige commerciële insteek. Lees er hier meer over. Lees ook ons privacy statement. Sommige inhoud (video's, iframes, formulieren,...) op deze website zal pas zichtbaar zijn na het accepteren van de cookies.

Accepteer cookies