6584.1:
The smallest surface supported ferromagnetic nanostructure – synthesis, reading and writing
Abstract
The future of magnetic information storage on hard disks or in MRAM's will critically depend upon our possibility to synthesize and individually address ferromagnetic entities which will be much smaller than the bulk superparamagnetic limit of about 10^5 atoms per bit. Grains of current hard disks have about this size. However, due to their size and shape distribution about 100 grains have to be used for a bit to achieve the required signal to noise. Thus currently a single bit of information consists of 10^7 atoms.
The first goal of our project is to synthesize via self-organized growth the smallest surface supported ferromagnetic nanostructure which retains its magnetic orientation at room temperature and possibly provides the bit unit in future high density magnetic storage applications. A second aim of the present project is to demonstrate individual reading of ferromagnetic units for densities of 10 Terra dots/in2 by means of spin-polarized STM. We also intend to individually write the magnetic units by injection of a spin polarized current from the tip of an STM.
The present supplement foresees to strengthen the manpower dedicated to the project by one PhD student in order to guarantee that the specified goals are achieved. In addition, we now also address the density limits in MRAM's. We do this by investigating the properties of magnetic nanostructures in spin valve systems. These are sandwiches formed by three layers, ferromagnet/oxide/ferromagnet. The relative orientation between the magnetizations in the two ferromagnets determines the state of the memory. Read-out is done by the transport properties across the sandwich depending sensitively on whether the two ferromagnets are parallel or antiparallel. Writing is currently done by magnetic fields created by write currents injected into two wires crossing at the addressed cell. The study of fm/oxide/fm systems has two advantages. First it is relevant for the second class of non-volatile magnetic memories, namely MRAM's, and second it approaches us to the reading and writing process of the individual nanostructures since the underlying mechanisms in spin-valves are identical to the one proposed in the original project for reading and writing of individual bits with spin polarized STM,.