Heat Assisted Magnetic Recording (HAMR) has often been discussed as a promising approach to push magnetic recording density to multiple terabit per square inch. As HAMR research is multi-disciplinary in nature, requiring expertise across several disciplines, such as optical, thermal and magnetic research, a system level simulation model is essential for studying the design trade-offs for each area. DSI researchers have designed a simulation model capable of investigating component level and system level performance of HAMR. The article below further describes the measures researchers have put in to make the simulation as close to a real case scenario as possible.



In magnetic recording, owing to the so-called superparamagnetic effect, the density of conventional recording is limited at around 1Tb/in2. Heat Assisted Magnetic Recording (HAMR) is a most promising approach to push magnetic recording density to multiple Tb/in2. In this technology, a near field optical head with nano-size beam output is used to heat up the magnetic media locally so as to reduce the media material’s coercivity, which makes the high coercivity media writable. Therefore, HAMR research is a multi-disciplinary research across optical, thermal and magnetic disciplines. In order to investigate component performance at the system level and provide an understanding of design trade-offs, a system level simulation model is essential.

Researchers at DSI have built the capability in system simulation for HAMR. The model covers near field optical head design and simulation by finite differential time domain (FDTD) method, media thermal profile distribution simulation, media magnetic written pattern simulation and read-out signal simulation by micro-magnetic method and the performance (signal to noise ratio, SNR) evaluation. The block diagram of the simulation is shown in Fig.1.
  

   


In the real recording system, the space between transducer and recording media top surface will be a few nanometers. For such a short distance, the media properties will definitely affect the transducer performance. Therefore, both the media material and structure are taken into consideration in our model. To make the simulation result closer to a real case scenario, the optical properties of the thin films used in the structure are measured. In order to obtain the thermal properties of the thin films, related capabilities are built and the thermal conductivity of the thin films is characterized.

  
   
In the case of c-aperture and FePt media, the optical distribution on the top surface of the recording layer is shown in Fig.2, and the optical spot size is 14 by 17nm2. With this optical distribution, the relationships between thermal spot size and in-plane thermal conductivity (kII) of the FePt film at different boundary thermal resistance between FePt layer and heat sink layer (Au with thermal conductivity of 180K/m/W) is shown in Fig.3. The requirements for thermal spot size at different recording densities are also plotted in the Fig.3. It was found that large in-plane thermal conductivity causes big thermal spot size and a large boundary thermal resistance results in considerable increase in the thermal spot size too.

In order to understand the possibility of achieving 4Tb/in2 recording density with required thermal spot, the thermal distributions under condition of kII=15W/m/K and BTR=1e-9m2K/W in different temperatures are used for magnetic writing simulation. The written patterns are shown in Fig. 4. The dependences of signal to noise ratio (SNR) on bit length at different temperatures are plotted in Fig.5 and it can be seen that there is an optimized temperature to obtain the largest SNR. In this condition, the recording density is limited at around 2.5Tb/in2. To further increase the density, optimizations of media, magnetic head and writing condition are needed.

With this system simulation, the component level optimization and system performance investigation can be carried out.
  
   

 
   


   
   


   


  Back To Top Next Page >