next up previous contents
Next: 1. Motivation Up: Scalable Parallel Micromagnetic Solvers Previous: Contents   Contents

Introduction

The effects of magnetism and magnetic materials have been exploited since the invention of the magnetic compass for navigation in the 10th century in China. It took until the 13th century, that this commercially and strategically important tool became known in Europe, and Charles Augustine de Coulomb made the first quantitative investigations only in the 18th century. However, after the pioneering work of Ampère and Faraday and the formulation of the theory of electrodynamics by Maxwell many new applications emerged.

The progress in the field of permanent magnets is very well illustrated by the maximum energy density product $(B H)_\mathrm{max}$. By the end of the 19th century magnetic steels with a $(B H)_\mathrm{max} \approx 2 \mathrm{kJ/m^3}$ were available. AlNiCo precipitation hardened magnets, which were discovered by Mishima in 1931, lead to energy density products as high as $90 \mathrm{kJ/m^3}$ by 1955. After the second world war hardferrites (ceramic oxides) were developed, and they are still commonly used because of the great abundance of their raw materials and low price. A major breakthrough was the discovery of magnetocrystalline anisotropy in rare-earth intermetallic compounds in the 1960s. Strnat et al. [1] found, that the combination of the high magnetic moment of iron and cobalt together with the high magnetocrystalline anisotropy caused by rare-earth elements gives permanent magnetic materials with excellent properties and energy density products of up to $90 \mathrm{kJ/m^3}$. Especially SmCo based materials retain a high magnetic ordering temperature, which makes them suitable for high temperature applications. Therefore, the HITEMAG project [2], whose aim is the development and optimization of permanent magnetic materials for application temperatures of up to $500 ^\circ \mathrm{C}$, concentrated on SmCo based precipitation hardened materials. As part of this project micromagnetic simulations have been carried out and the results are presented in Sec. 7.

The highest energy density products so far have been obtained with rare-earth iron based permanent magnets. In 1984 Sagawa et al. [3], Croat et al. [4] and Hadjipanayis et al. [5] obtained energy density products of up to $300 \mathrm{kJ/m^3}$ for a material based on Nd$_2$Fe$_{14}$B. In the following years continuing improvement of the production route has resulted in energy density products in excess of $400 \mathrm{kJ/m^3}$ [6].

A very similar development can be observed in the area of magnetic recording. In 1956 IBM introduced the 305 RAMAC (Random Access Method for Accounting and Control) with a capacity of 5 MB. It was the first magnetic storage device, which stored digital information by writing magnetization patterns on a ``hard disk'' with a thin film of granular magnetic material instead of magnetic tapes. One of the major advantages was, that any position on the disk could be directly accessed by the read/write heads and it was not necessary to wind a tape any more. Since then many new discoveries and developments have improved computer hard disks.

The areal density determines the amount of information, which can be stored on a given area of a hard disk. This figure of merit measures the performance of hard disk media in a similar way like the energy density product for permanent magnetic materials discussed above. And it has shown a similar development during the last 20 years.

In 1985 the typical areal density in mass production was $20 \mathrm{MB/in^2}$ (megabits per square inch). The industry trend showed a typical increase of about 27 % per year. In 1992 it exceeded $100 \mathrm{MB/in^2}$ and the annual growth rate jumped to approximately 50 % per year. Starting with the introduction of giant magnetoresistive read-write heads in 1997 the areal density has been doubling every year [7]. Present-day drives have an areal density of about $30 \mathrm{GB/in^2}$ and read-write heads based on the extraordinary magnetoresistance effect will allow data densities beyond $100 \mathrm{GB/in^2}$ [8,9].

As an alternative to these thin film granular magnetic storage media ferromagnetic nanostructures are considered for the basic information storage elements in magnetic random access memories (MRAM), high density magnetic storage media, and magnetic sensors. These structures can be produced using well established techniques for semiconductors but they have several advantages over today's semiconductor based materials including nonvolatility, nondestructive readout, radiation hardness, low voltage, and unlimited read and write endurance [10,11]. Therefore, the magnetic properties, switching behavior, and switching dynamics of magnetic nanostructures are of great interest, and the results for cylindrical nanodots and elliptical nanoelements are presented in Sec. 9 and Sec. 10, respectively.

Both research areas, permanent magnets and magnetic storage media, have made tremendous progress during the last decades and they have become key technologies in today's information society. In order to keep the pace of these developments and push the limits, many research projects are carried out worldwide and this thesis is one small contribution.


next up previous contents
Next: 1. Motivation Up: Scalable Parallel Micromagnetic Solvers Previous: Contents   Contents
Werner Scholz 2003-06-08