Overview
The Magnetic Random Access Memories (MRAM) group develops advanced concepts in this emerging technology. The goal is to realize cells with improved thermal stability, lower power consumption and/or faster switching. Our research covers material stack deposition, nano-fabrication and electrical test evaluation, for applications as standalone memory and non-volatile logic and more recently in neuromorphic computing architectures.
Research directions
Perpendicular Anisotropy Materials
High energy barriers for spin transfer torque (STT) MRAM cells can be achieved with perpendicular anisotropy magnetic tunnel junctions. Solutions for high density MRAM cells to diameters below 20nm require continuous improvements in perpendicular surface anisotropy, while maintaining high TMR properties.
Perpendicular STT MRAM
Evaluation of MRAM concepts requires simulation of expected reversal mechanisms and electrical characterization of individual cells. We aim at understanding dynamics of magnetization reversal and the expected impact of stack modifications to explore application specific optimizations.
Nanofabrication Challenges
Innovation on dense MRAM using pre-patterned substrates, CMOS integration of multifunctional cells and sub-10nm lateral sizes. Tunnel junction nanofabrication in our platform is essential to evaluate MRAM concepts and performance.
Perpendicular Shape Anisotropy
A solution for sub-10nm cell sizes uses high aspect ratios to generate perpendicular shape anisotropy providing scalable retention at the smallest cell sizes. Spin transfer torque switching is possible in these cells, where the reversal dynamics is now under study.
The team
Former members
Post-docs
- Andrey TIMOPHEEV (2014-2017)
- Van Dai NGUYEN (2016-2018)
- J. Ranier Roiz (2015-2016)
- Nikita Strelkov (2016-2019)
PhD
- Luc TILLIE (2015-2018)
- Nicolas PERRISSIN (2015-2018)
- Jyotirmoy CHATTERGEE (2014-2017)
- Hieu Tan NGUYEN (2013-2016)
- Antoine Chavent (2013-2015)
Process Engineers
- Jude GUELFFUCCI (2015-2017)
- Nathalie LAMARD (2016-2017)
- Guillaume LAVAITTE (2015-2016)
Projects
- Samsung SGMI (2014-2017)
- ANR Excalyb (2014-2017)
- Heumem (2015-2018)
- EU-FET Spice (2016-2019)
- EU Great (2016-2019)
- ERC Magical (2015-2020)
Partners
- CEA LETI, Grenoble, France
- Institut NEEL, Grenoble, France
- Crocus Technology, Grenoble, France
- Samsung, San Jose, USA
- Singulus AG, Kahl am Main, Germany
- Aarhus University, Aarhus, Denmark
- Radboud Universiteit, Neijmegen, Netherlands
Recent news
- WP4 : DESIGN OF LOW-POWER HYBRID CMOS/MAGNETIC CIRCUITS (July 02nd, 2015)
Within HYMAGINE, circuits of increasing complexity have been conceived from simple non-volatile logic gates to microcontrollers or microprocessor. Below is an example of magnetic Look-Up-Table (MLUT) conceived within HYMAGINE and an example of hybrid CMOS/MTJmicroprocessor. Magnetic LUT ... - WP2 : SWITCHING SPEED AND COHERENCE (July 02nd, 2015)
Thanks to their unique set of assets (non-volatility, speed, density, endurance), STT-MRAM are seen as a unique candidate for DRAM and/or Cache SRAM replacement allowing to drastically reduce the power consumption of electronic circuits thanks ... - WP3 : MODELLING AND DESIGN TOOLS (July 02nd, 2015)
Modelling and design tools were developed in the frame of HYMAGINE to cover both the fundamental and design aspects of the project. Concerning the fundamental aspects, we developed a code allowing to calculate both the transport ... - MAGNETOSTATICS OF SYNTHETIC FERRIMAGNET ELEMENTS (July 02nd, 2015)
Olivier Fruchart,-, Bernard Diény We calculate the magnetostatic energy of synthetic ferrimagnet (SyF) elements, consisting of two thin ferromagnetic layers coupled antiferromagnetically, e.g. through RKKY coupling. Uniform magnetization is assumed in each layer. Exact formulas as ... - EMERGING NON-VOLATILE MEMORIES: MAGNETIC AND RESISTIVE TECHNOLOGIES (July 02nd, 2015)
B Dieny In 2010, the International Technology Roadmap for Semiconductors (ITRS) published an assessment of the potential and maturity of selected emerging research on memory technologies. Eight different technologies of non-volatile memories were compared (ferroelectric gate ...
