Go to main content
[Group of people]

School of Computer Science Intranet

Manchester Patterned Media Research

What's Patterned Media?

In a nutshell ...

3D read head/media geometryConventional magnetic storage media are composed of a continuous magnetic thin film to which digital data is recorded using a magnetic write head that flies in close proximity to the surface of the magnetic film as the disc spins. As the demand for high-density digital storage capacity grows, magnetic hard disc drives have met the these demands by the continued increase in the area storage capabilities of continuous media, via the reduction in recorde mark size and the narrowing of track widths.

Current commercial magnetic disk drives offer storage densities in excess 100Gbits/in2, and are expected to offer storage capacities approaching 1Tbit/in2 using perpendicular recording media. However, it is difficult to foresee how far the present storage paradigms can be extended to meet increasing demand; due to well-known fundamental physical limits that restrict the storage densities of current technologies. Readout signal-to-noise ratios are determined by many factors, including transition noise and bit shifts due to strong inter-granular magnetic coupling and in particular, the number of grains in a recorded domain. In order to increase storage density the size of the recorded domains must be reduced, however this result in a need to reduce the size of the magnetic grains that form the recording layer in order to maintain sufficient signal-to-noise ratio. However, a reduction in grain size can result in thermal magnetic instability, the so-called superparamagnetic limit. A solution is to increase the medium coercivity, however, this makes it difficult to write information to the medium due to the high write fields required.

Patterned Media One solution to this problem of ultra-high-density recording on continuous media is the use of a patterned recording medium. Here, the continuous magnetic thin film is patterned to form an array of nanometre sized, isolated, magnetic islands, each storing a single bit of information. In order to attain a storage density of 1Tbit/in2 each island occupies an area of approximately 25nm x 25nm ... very very small (1/2000 the width of a human hair). Since each island acts as a single bounded domain, the grains are strongly coupled; consequently the grain size need not be reduced. Furthermore, the effects of magnetic coupling between recorded marks, and hence transition noise, are effectively reduced resulting in an improved signal-to-noise ratio.

A nice pictorial representation of patterned media with respect to conventional media can be found on the Hitachi Global Storage Technologies website.

There are a number of problems associated with the development of patterned media as a viable storage medium, the most important of which, is how do you fabricate small islands (typically 10-20nm in diameter) using the fabrication techniques currently available.

Manchester Patterned Media Research

The EISS research group have worked on the development of patterned media for nearly 10 years, past and present projects include:

Nanoscale Magnetic Storage: To 65 GBits/sq.in. and beyond

C D Wright, E W Hill and J J Miles, EPSRC GR/L83462/02, Oct. 1999 - Sept 2002, £97,115.

Abstract

This programme will investigate, both theoretically and experimentally, magnetic data storage on patterned, nanoscale media to densities of 65 Gbits/sq.in. and beyond. A micromagnetic model of the switching, stability and read/write behaviour of a variety of patterned media with a range of microstructures and anisotropies will be produced and used to optimise material performance. A theoretical study of the important thermal stability problem in such media will be carried out, leading to a combined micromagnetic and thermal simulation to allow for optimisation of the storage process. Recording films will be prepared by our collaborators at CEA in Grenoble and patterning will be carried out primarily at Manchester using e-beam lithography. Experimental facilities will be developed to enable the writing and reading on such patterned media. Both near-field and MFM write and read techniques will be developed.

Other staff/students involved

  • Paul Nutter (Research Fellow)

Analysis of Data Recovery Techniques for use with Patterned Media Storage

P W Nutter and B K Middleton, EPSRC GR/R63479/01, Sept. 2002 - Sept. 2005, £68,019.

Abstract

In the quest for reliable ultra-high-density media for data storage applications, patterned media is clearly recognised as providing a solution to the problems associated with continuous media, promising potential storage densities far exceeding those attainable using conventional technologies. This proposal is highly novel and extremely timely. We aim to investigate the problem of reliable data recovery from patterned media and the implications of the discrete media upon the performance of the data recovery process; an area where there is little information in the public domain. The data recovery process must be fully developed and characterised in order to unleash the full potential of patterned media as a viable storage solution. Read channel designs will be investigated, through extensive simulation and analysis, for use in magnetic and magneto-optic storage systems incorporating patterned media with perpendicular magnetic anisotropy. Initial designs will be established through the analysis and application of established data recovery techniques and new schemes will be developed that are sympathetic to the discrete nature of the media. Read channel performance will be investigated through the analysis of bit-error-rate performance in the presence of a variety of noise sources, in particular, those associated with the discrete nature of the media, determined through experimental replay analysis from sample patterned media.

Other staff/students involved

  • Ioannis Ntokas (PhD Student) - Simulation of channel designs for patterned media storage
  • Matthew Manfredonia (PhD Student) - Simulation of pattered media in near-field optical storage systems
  • Branson Belle (PhD Student) - Patterned media fabrication

Further information about this project can be found here.

Nano-Patterned Storage Media: Noise & Density Limits

J J Miles, E W Hill and P W Nutter, EPSRC EP/E017657/1, 2007 - 2010, £652,551.

Abstract

Hard disk drives have continued to increase in capacity and decrease in cost per bit stored since their invention 50 years ago, with the number of bits stored per unit area increasing by a factor 100 million. This remarkable development was achieved without fundamental change in the storage paradigm, but recently more change has become necessary to achieve continued growth in performance, and modern disks now store information in magentisation perpendicular to the disk surface rather than in the plane. This change is expected to allow continued growth in capacity for around five years, beyond which a more fundamental change will be needed. The most likely alternative is thought to be 'Patterned Media', in which the surface of the disk is manufactured as discrete islands of magnetic material, each island storing one bit of information. A number of research groups have started to investigate different methods of manufacture of such materials, from direct e-beam writing to self-assembled templates. Although the fundamental parameters of some materials have been revealed by microscopy and magnetometry, and some basic storage experiments have been performed, it has not yet been possible to establish the comparative merits of different fabrication processes. We aim to develop an instrument capable of measuring the storage capabilities of experimental samples of such nano-structured materials and to use it to study the characteristics of a range of materials made in house and from other laboratories. By comparing these measurements with model prodictions we aim to uncover relationships between the fabrication method and the nature of noise, and to determine the requirements that the system performance places upon the fabrication process. Such understanding will reveal the ultimate limits of the technology.

Other staff/students involved

  • Branson Belle (Research Asscoiate) - Patterned media fabrication/Instrument development
  • Marios Alexandrou (PhD Student) - Patterned media characterisation
  • Josephat Kalezhi (PhD Student) - Write modelling in patterned media storage systems
  • Yuanjing (Lizzy) Shi (PhD Student) - Modelling of channels for nano-island data storage systems

Created by Paul Nutter