System and Media Optimizations for improved HAMR Performance

Abstract

It is said that data never sleeps. It has a ubiquitous presence and is being generated at an exceptional pace from different sources. The insurmountable demand for data must be met with an equally fast paced data supply which has led to the development of data servers by companies like Microsoft and Google. In spite of facing an existential crisis due to the development of SSD’s, the Compound Annual Growth Rate (CAGR) of ~ 40% maintained by the HDD industry has ensured that these devices are a necessity when it comes to large scale data storage. The growth of the HDD industry is helped by the fact that a huge amount of research is dedicated to developing new data recording technologies to improve the storage capacity of HDD’s. Significant investment has gone into developing a new data recording technology called Heat Assisted Magnetic Recording (HAMR) that is expected to improve the storage density up to at least 5Tb/in2. In conjunction with other improvements (like the development of Bit Patterned Media BPM), the expected density output for HAMR can be even higher. The optimization of the HAMR technology has focused on different aspects of the HAMR system, optical, magnetic, mechanical and electrical. In this thesis, different system and media optimizations that may help improve the HAMR performance are explored. The effect of doped L10 FePt media, which is an extremely popular HAMR media, is considered to understand how a change in its Curie temperature (Tc) can actually influence its intrinsic magnetic properties. This is followed by the implementation of micromagnetic simulations with the use of the stochastic Landau Lifshitz Gilbert (LLG) equation that is used to mimic the magnetization dynamics of grains. These simulations are used to vary the media properties and HAMR process parameters to optimize a thin 3nm (write layer)/6nm (storage layer) Thermal Exchange Coupled Composite (ECC) HAMR media. Introducing finite exchange coupling between the grains of the write layer and scaling the damping in the write layer are techniques that can help reduce the DC noise and improve the Signal to Noise Ratio (SNR). The Ensemble Waveform Analysis technique identifies Transition SNR as the main cause of SNR variation. This optimization process lends credence to the idea that a thinner composite media may be used to realize significant enhancements of SNR. Micromagnetic simulations are also used to address an important issue related to HAMR; the high temperature for writing data can cause heating issues with long term HAMR use. A low temperature Thermal ECC media is proposed that can significantly reduce the writing temperature (by about 34%) and that can reduce the peak temperature of the heat spot used to heat the media in the HAMR process by 200K. This is followed by an analytical formulation that is derived to calculate the transition jitter in the HAMR process. The jitter is known to depend on the grain size as well as the heat spot thermal gradient. It also depends on the Voronoi Grain Size Distribution, as well as exhibiting a surprising nonlinear dependence on the reader width. By combining the noise due to these dependencies the analytical formulation can be derived. This simple formulation provides both physical insight and conserves computational time relative to lengthy (and complex) recording simulations. A detailed analysis of the Adjacent Track Erasure (ATE) in the HAMR process is also explored. The numerical extent of ATE in different HAMR media is established and techniques are implemented using micromagnetic simulations in an attempt to reduce the ATE effect. A hypothesis is established to explain the presence of extent of ATE in different HAMR media. Research in the area of HAMR process and system optimization is of huge importance especially since the data storage industry has invested a lot in terms of research and manpower in this technology. Potential directions of research include techniques to reduce the ATE, improving the designs of different HAMR system components and developing better data post processing techniques like Neural Networks and 2D detectors.