CD-Labor Advanced Magnetic Sensing and Materials

Within this project we aim to apply advanced simulation techniques for the development and design of new sensor devices for the automotive industry, which are also applicable for bio application. This project which involves world renowned experts in micromagnetics and Infineon, the world market leader in semiconductor based sensors will lead to the unique opportunity to solve the challenging problem of computer aided design and development of large scale semiconductor-based magnetic sensors.

The applicant’s micromagnetic simulation software FEMME is used by major companies worldwide for the optimization of giant magnetoresistance (GMR) and tunnel magnetoresistance (TMR) read heads in hard disc drive systems. The application of this software tool for the design of magnetic sensors for automotive industry and bio application is far from obvious and it is a very challenging task. The challenge is that the involved lateral dimensions in this new application area are at least by a factor 10 larger. This increases the number of unknowns for 3D simulation by a factor of 1000. Hence, innovative and new modeling concepts have to be developed in order to succeed with this task.

We aim to develop a flexible 3D finite element package which can be executed highly parallel in order to make use of the power of recent multicore CPU’s. We will achieve this task by formulating the stray field problem which arises in micromagnetics in an integral form and apply efficient compression techniques (H-matrices) in parallel to reduce the computational time. The method to calculate equilibrium states will be optimized for the application. We aim to apply efficient energy minimization routines for finding the magnetic state as function of the external field, which varies in the kHz regime.

This advanced micromagnetic software will be used for (i) understanding the magnetization processes of state of the art GMR sensors (ii) design of new functional structures (iii) as an input for a device models, which are required for electronic circuit simulations.

We aim to develop device models using the following complementary approaches. (i) Bottom / Up: This design starts from accurate simulations and tries to reduce the complexity. Here we aim to use a new approach that reduces the computational complexity by performing simulations in eigenmode basis. (ii) Top / Down: Device models that obtain the input from experiments and micromagnetic simulation. This input data determine the parameters of macroscopic hysteresis models.

A particular design problem which will be solved within this project is the so called “jitter-problem” which significantly reduces the accuracy of state of the art sensors for speed measurement. Later in the project completely new sensor concepts based on magnetoimpedance and magneto – acoustic sensors will be investigated for the mentioned and future application.
The work of this proposal brings together Austrian expert groups which are world leaders in their field. This interplay between simulation and experiments is believed to be the key success factor of this project. This combined effort will be used to optimize existing sensors, to solve performance issues, and to develop new designs for improved sensing.