A statistical model of write-errors in bit patterned media

摘要

In bit patterned media (BPM), a magnetic data storage medium is patterned into nanoscale magnetic islands each representing one binary digit (bit). The recording performance of BPM depends upon the variability of island position, geometry, and magnetic characteristics. To understand the impact of the distributions of these parameters on the performance of BPM a detailed statistical model of write-errors has been developed. The islands can either be single layer or two-layer exchange coupled composite (ECC) structures. The modeling of ECC islands was made possible by the development of a 2-spin model to calculate switching field, coercivity, and energy barrier as a function of applied field which enables medium design to be optimized for a non-uniform write head field. The statistical model predicts coerciviries and error rates in good agreement with a micromagnetic model but at significantly lower implementation and computational cost and shows good agreement with experimental data from drag-test experiments at 500 Gbit/in2. The model enables the position tolerance of the write head to be determined from the magnetic characteristics of the write head and the storage medium and it is therefore a valuable system design tool.