A portion of the disclosure of this patent document contains material which is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure, as it appears in the Patent and Trademark Office patent file or records, but otherwise reserves all copyright rights whatsoever.
This invention relates to high density data storage.
Software developers continue to develop steadily more data intensive products, such as ever-more sophisticated, and graphic intensive applications and operating systems. As a result, higher capacity memory, both volatile and non-volatile, has been in persistent demand. Add to this demand the need for capacity for storing data and media files, and the confluence of personal computing and consumer electronics in the form of portable media players (PMPs), personal digital assistants (PDAs), sophisticated mobile phones, and laptop computers, which has placed a premium on compactness and reliability.
Nearly every personal computer and server in use today contains one or more hard disk drives (HDD) for permanently storing frequently accessed data. Every mainframe and supercomputer is connected to hundreds of HDDs Consumer electronic goods ranging from camcorders to TiVo® use HDDs. While HDDs store large amounts of data, they consume a great deal of power, require long access times, and require “spin-up” time on power-up. Further, HDD technology based on magnetic recording technology is approaching a physical limitation due to super paramagnetic phenomenon. Data storage devices based on scanning probe microscopy (SPM) techniques have been studied as future ultra-high density (>1Tbit/in 2) systems. Ferroelectric thin films have been proposed as promising recording media by controlling the spontaneous polarization directions corresponding to the data bits. However, uncontrolled switching of the polarization direction of a data bit can undesirably result in ferroelectric thin films as data bit density increase.
Further details of the present invention are explained with the help of the attached drawings in which:
Ferroelectrics are members of a group of dielectrics that exhibit spontaneous polarization—i.e., polarization in the absence of an electric field. Ferroelectrics are the dielectric analogue of ferromagnetic materials, which may display permanent magnetic behavior. Permanent electric dipoles exist in ferroelectric materials. One common ferroelectric material is lead zirconate titanate (Pb[ZrxTi−x]O3 0<x<1, also referred to herein as PZT). PZT is a ceramic perovskite material that has a spontaneous polarization which can be reversed in the presence of an electric field. PZT can be doped with either acceptor dopants, which create oxygen (anion) vacancies, or donor dopants, which create metal (cation) vacancies and facilitate domain wall motion in the material. In general, acceptor doping creates hard PZT while donor doping creates soft PZT. In hard PZT, domain wall motion is pinned by impurities thereby lowering the polarization losses in the material relative to soft PZT, but at the expense of a reduced piezoelectric constant.
Referring to
Ferroelectric films have been proposed as promising recording media, with a bit state corresponding to a spontaneous polarization direction of the media, wherein the spontaneous polarization direction is controllable by way of application of an electric field. Ferroelectric films can achieve ultra high bit recording density because the thickness of a 180° domain wall in ferroelectric material is in the range of a few lattices (1-2 nm). However, it has been recognized that maintaining stability of the spontaneous polarization of the media may be problematic, limiting use of the media in memory devices.
Referring to
A system is stable, in a macropscopic sense, when the characteristics of the system do not change with time but persist indefinitely. The stability of a system can be approached if the free energy of the system is at a minimum for a given combination of temperature, pressure and composition. The free energy of a system comprising a media including a ferroelectric layer can be approximated by equation:
G=GO+U
wherein Gois a part of the free-energy attributable to a non-zero polarization, and U is a part of the free-energy that is not related to the polarization, and which can be substantially attributed to depolarization energy.
The depolarization energy, U, is negligible where the polarization is small;; however, the polarization of perovskite ferroelectric crystals such as PZT is relatively large. A ferroelectric layer comprising a single domain can result in a large depolarization field. The depolarization field can be expressed by the equation:
wherein ε* is the effective permittivity, Po is the polarization, V is the domain, d is the domain width, and t is the domain thickness. The depolarization energy is reduced by breaking the ferroelectric layer into domains of different polarization, which consequently results in domain walls having domain wall energy Uwall that contribute to the free energy of the system so that the free energy of the system is approximated by equation:
G=Go+Uwall+Udepolarization
The domain wall energy Uwall can be expressed by the equation
wherein σ is the domain energy per area.
However, where the media has an asymmetric structure, a hypothetical energy diagram of a domain of a ferroelectric layer plotting energy, G, as function of polarization is asymmetric and can resemble the hypothetical energy diagram of
The asymmetric relationship of polarization energy and ferroelectric-to-paraelectric transition energy can result in undesirable influences of neighboring domains on one another. For example, where an up domain has a relative lower ferroelectric-to-paraelectric transition energy comparable to a down domain, the up domain can be said to be more stable than the down domain for a given domain size. If the up domain and down domain is formed having an identical size, the more stable up domain can flip the polarization of a portion of the down domain to the polarization of the up domain. The up domain can influence the down domain to expand in size and consequently reduce the down domain in size. This interaction can halt where equilibrium is reached as wall energy of the down domain increases as a result of decreasing domain size. However, it is possible that the entire down domain can be flipped by the neighboring up domain, resulting in lost information.
Embodiments of media and methods in accordance with the present invention can be applied to improve stability of domain polarization in ferroelectric-based probe storage devices, thereby improving data retention. It should be noted that in some contexts, domain can refer to a discrete unit such as a data bit comprising material having non-uniform dipole orientation. However, as used herein, domain refers to a volume of a ferroelectric material having uniform dipole orientation and defined by domain walls. As used herein, a data bit refers to a discrete unit of information and can comprise one or more domains.
In an embodiment, a media and method of improving data retention for ferroelectric-based probe storage devices can comprise arranging domains within a media to obtain a macroscopically minified free energy. Domains can be arranged in groups of two or more domains, a group representing a data bit. The number of domains grouped together to form a data bit can depend on the energy characteristics of the media and the screening charges formed on the surface of the media. For example, for a media having energy characteristics as reflected in the energy diagram of
Grouping of domains can be adjusted to suit the energy diagram of the ferroelectric layer of a media, which as noted above can depend on domain thickness, domain width, properties of the ferroelectric material, and other parameters. For example, if a media has a hypothetical energy diagram as shown in
As will be appreciated upon reflecting on the current teachings, an adjacent track (also referred to herein as flanking track) can influence a minimum free energy (and therefore stability) of the track to which it is adjacent, just as domains adjacent within a track can influence a stability of one or both of the domains. Tracks (and domains within tracks) can be written to achieve a desired free energy to result in a desired stability across tacks. In alternative embodiments adjacent tracks can be spaced to reduce instability across adjacent tracks. Alternatively, as shown in
Identifying data bits as groupings of an up domain with a down domain can further controllably limit undesirable arrangements of domains across a track. For example, where a track comprises in part a string of data bits “00000001111111,” the grouping of up and down domains allows recovery of a clock signal, despite a long run of “0” data bits and a long run of “1” data bits. Across track arrangement of data bits can further improve stability. For example, some embodiments of coding schemes can arrange data bits so that the smaller of the up domain and the down domain is not positioned adjacent to more than one identically polarized domain in the tow adjacent tracks.
Grouping of domains can be adjusted to suit a combination of the energy diagram of a media and general screening charges to account for total free energy. The free energy characteristics of a down domain relative to an up domain cannot be easily calculated. However, the ratio of up domains to down domains and an approximation of general screening charges and defects can be experimentally determined for providing a free energy for relatively stable domains at given conditions, wherein the conditions can include ferroelectric considerations and environmental conditions, such as thermal effects. To experimentally determine a desired ratio, up and down domains having different ratios can be written to the media for certain media conditions (e.g., screening ratio, ferroelectric layer thickness, degree of asymmetry). Temperature-accelerated testing can be performed on the media, and a comparison drawn of the ratios of up and down domains to judge the desired ratio (i.e., the most stable and/or most preferred ratio).
In some embodiments, stability of domains can be further improved by arranging data bits to provide a desired balance of data bit states.
In alternative embodiments, a data bit can comprise a single domain. For example, a “0” can be represented by one of an up domain and a down domain and a “1” can be represented by the other of the up domain and the down domain. The data bits can be coded to best approximate a stable ratio of up domains to down domains. Software can be employed to keep track of the arrangement of data. Such schemes are know in the art for ensuring clock recovery for data streams. A useful scheme can group blocks of data using an algorithm to achieve an arrangement that achieves a ratio criterion approaching a minified total energy of the system (e.g., 66:33, 40:60).
In still further embodiments, data bits can be represented by a single domain, thereby increasing maximum density. To achieve a minified total energy, a media can be divided into sectors. Referring to
A minimum possible sector size can depend on the characteristics of the ferroelectric layer. As instability of one of the up domains and down domains becomes more problematic, it may be desired that sector size be relatively small. As shown in
In still other embodiments, coding techniques can be applied to scramble data within a single block or multiple blocks to achieve information streams that result in a desired ratio of up domains to down domains. Data can be scrambled to assure that each bit is independent, or equally likely, within a channel. Scrambling can avoid continuous worst case patterns within the channel. In combination with an RLL code, scrambling allows shaping of the spatial and temporal spectrums to achieve improvements in data retention. An RLL code can force run length constraints with substantial certainty, thereby improving retention. Thus RLL code can be used with ferroelectric media to improve retention at very high densities. Such coding techniques can further take advantage of error correction code (ECC) applied when scrambling data to be written to a block. ECC is applied to meet density and reliability requirements.
In still further embodiments, a background pattern of polarization can be applied to the media, over which information can be coded. The background pattern can be devised so that the background provides stability, reducing the influence of neighboring bit. For example, as shown in
As shown in
The foregoing description of the present invention has been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Many modifications and variations will be apparent to practitioners skilled in this art. The embodiments were chosen and described in order to best explain the principles of the invention and its practical application, thereby enabling others skilled in the art to understand the invention for various embodiments and with various modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the following claims and their equivalents.