SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : C-Cube -- Ignore unavailable to you. Want to Upgrade?


To: John Rieman who wrote (50610)1/24/2001 11:48:04 AM
From: BillyG  Read Replies (1) | Respond to of 50808
 
DVD recording technology.....
nikkeibp.asiabiztech.com

DVD Researchers Look Beyond 50 Gbits/Inch2 Density

The development of technology for next-generation DVD optical disks is
approaching its climax, and proposals are being submitted for physical
formats. The focus of R&D is now shifting to the next generation, and
areal recording density is expected to jump once again beyond that of
hard disk drives.

The next generation of digital video disk (DVD) technology is being prepared
for full-scale commercial application in 2002 or 2003. A number of equipment
manufacturers, having been steadily solving the various problems with
constituent technology, are now submitting physical format proposals as the
groundwork for standardization of the next generation of optical disks. There
has been a flurry of announcements of technologies that slash
volume-production cost.

At the same time, the development of optical technology for drives and media
to be released even further on is also picking up steam. The new optical
technology is expected to offer areal recording densities which exceed that of
hard disk drives (HDD), from 50 to several hundred Gbits/inch 2 .

Technical proposals for next-generation DVD optical disks can be broadly
classified into three groups by disk structure: (1) disks covered with recording
films, which are covered with a thin (about 0.1mm) layer; (2) sandwiched
media made of two disk substrates bonded together, each 0.3-0.4mm thick;
and (3) media of the same structure as existing DVD disks, with a substrate
thickness of 0.6mm.

DVR-Blue Essentially Complete

Equipment manufacturers who have been announcing new developments in
types (1) and (2) are now submitting proposals for physical formats (Fig 1).

A group composed of Philips Research Laboratories and Sony Corp has
disclosed the most detailed set of specifications. Many engineers feel that the
"DVR-blue" technology has matured almost to its final stages. It uses
phase-change media with a blue-violet laser as the light source.

The basic format structure is the almost the same as the "DVR-red"
specification presented by the pair at the International Symposium on Optical
Memory/Optical Data Storage (ISOM/ODS) '99 in July 1999, using a red
laser light source. It uses land-groove recording, defining eight header regions
per track to store address data and information for tracking servo correction.
These headers are arranged in a row from the inner to outer tracks,
segmenting the disk into eight regions like the spokes of a wheel.

Each recording track, defined as the region between two adjacent headers, is
made up of multiple sectors, with addresses being managed on a system based
on the leading header and the wobble number. Compared to DVD random
access memory (RAM) disks with one header per sector, format efficiency is
enhanced to about 90%. The recording capacity is probably fixed at 22.5
Gbytes.

Different Address Methods

Victor Co of Japan, Ltd, which has been developing technology for approach
(1), has proposed a physical format based on groove recording. The firm
selected the technology because it promises simple manufacturing and is easy
to assure compatibility with read-only disk formats.

Lands which are not used for recording data are split as prepit where address
signals are embedded. The single-side capacity for a 120mm disk is 17
Gbytes. It is more difficult to boost areal recording density than with the
land-groove recording scheme, but when the numerical aperture (NA) of the
object lens is improved from the current 0.7 - 0.78, the smaller beam spot will
make it possible to have 22 Gbytes per side.

Korean firms LG Electronics Inc and Samsung Electronics Co have both
announced technologies related to approach (2), although they have not
proposed them as actual formats. Both firms are working on landgroove
recording implemented by embedding address data with wobbling. This
scheme has been chosen because it is harder to boost single-side recording
capacity in approach (2) than in approach (1) with the thin cover layer.

25GB with 2-Layer Recording

At ISOM 2000, held in September last year, the industry showed
considerable interest in the activities of groups working on approach (3). Until
the function, those groups had kept their research and development results
quiet, but finally papers were presented. Hitachi, Ltd, Matsushita Electric
Industrial Co, Ltd and others actively offered proposals on next-generation
optical disks, and the two-layer recording film structure proposed by both
firms was especially interesting. Until these proposals, there had been no
reports on tests of two-layer disks using blue-violet lasers.

The total recording capacity (one side with both recording layers) was 25
Gbytes for the Hitachi disk, and 27 Gbytes for Matsushita's. This is equivalent
to recording about two-and-a-half hours of high-definition television (HDTV)
broadcasting. Both firms claim a carrier/noise (C/N) ratio of about 50dB for
both layers, when recording or playing at about 35 Mbits/s. Technically, this
means the C/N ratio is more than sufficient for practical use.

Matsushita developed a Ge-Sn-Sb-Te recording layer for the two-layer
media, and had already announced a Ge-Sb-Te recording layer for a
two-layer recording technology using red laser light. The shorter light
wavelengths, however, made it impossible to assure an adequate C/N ratio
and an erasure performance with a simple change of film thickness, and so the
new recording material was developed.

One of the key concerns was a loss in transparency of the recording layer at
shorter wavelengths, which made it necessary to thin out the first recording
layer to about 6nm. When the recording layer is made too thin, however, the
crystallization speed of Ge-Sb-Te recording layers drops off. To resolve this
problem, the firm added Sn to create Ge-Sn-Sb-Te, maintaining crystallization
time even with a thinner first recording layer.

Next-Gen ROM Manufacture

Another issue attracting considerable interest in the next-generation optical
drive field is volume-production technology, especially that which reduces
cost. Pioneer Corp has solved the basic technical problems involved in using
injection molding to press read-only disks with 25 Gbytes of capacity per side.
The firm uses the thin cover layer approach, and its technology raises the
possibility of enabling packaged media with HDTV resolution movies in the
near future.

The firm has used an injection molding system designed for existing DVD disks
to fabricate a 25-Gbyte read-only disk. Measured playback signal jitter (Fig
2) is kept under 10%. The firm believes it is now possible to make
next-generation optical media without major equipment modifications.

Lens Manufacture by Etching

There are also proposals for new volume-production techniques for optical
components. Sony, for example, has announced a plasma etching technique
for forming multiple lenses at once on a glass substrate. The firm is first aiming
to use this technology for the DVR-blue specification.

The specification uses a combination of large and small aspherical lenses to
create an object lens with an NA of 0.85. Glass aspherical lenses, however,
suffer from excessive metal mold costs, and low productivity because only a
few can be made at a time.

The firm's newly-developed technology is capable of batch-forming about 400
lenses on a single glass substrate two inches (about 50mm) in diameter (Fig 3).
The firm plans to eventually use an 8-inch (about 200mm) substrate to form
thousands of lenses at once.

The firm has only developed a prototype of the smaller of the two lenses, the
one closest to the disk. Sony plans to increase the lens diameter of the smaller
lens along with the working distance (WD: distance between disk and lens),
and apply the same fabrication method to the larger lens as well.

If both lenses are made with this volume-production technology, it has the
potential to significantly reduce optical pickup assembly cost as well. The glass
substrates, one with large lenses and the other with small lenses formed, can
be arranged vertically and semiconductor wafer technology (adopted for
positioning) can be used to batch-align all the lenses at once.

Near-Field Recording

Now that some of the key technical problems are being resolved for
next-generation optical disks, researchers are beginning to address the issues
of how to boost areal recording density past HDD levels. Already, densities of
50 Gbits/inch 2 , equivalent to that of existing HDDs, have been achieved in
experiments. At ISOM 2000, there were a number of reports of similar
densities under close-to-product conditions, and of significantly higher
densities as well. In particular, there were many reports on nearfield
technology.

For example, Sony has succeeded in recording and reading back data using
near-field technology with a solid immersion lens (SIL), while performing
tracking control. Near-field technology has finally taken a big step from the
demonstration of the recording and read principles to the investigation of
practical implementation in actual equipment. An optical disk was fabricated
with a land-groove structure, with tracking control data in the grooves, and
marks recorded to the land tracks (Fig 4). The control method was the
standard push-pull design.

The minimum recording mark length was 107nm, and record/read using those
marks showed a C/N ratio of 41dB. Areal recording density, as calculated
from the linear recording density, was equivalent to 40 Gbits/inch 2 . Track
pitch was 360nm, however - wider than the 180nm equivalent to that areal
recording density.

Super RENS Improves C/N Ratio

About ten papers have been presented on super-resolution near-field structure
(Super RENS), a type of near-field record and read technology, and the
research results themselves showed considerable progress. The National
Institute for Advanced Interdisciplinary Research of Japan, in co-operation
with Sharp Corp, has developed a structure that roughly doubles signal read
strength in Super RENS, by utilizing a physical phenomenon called surface
plasmons.

The record/read principle was announced at ISOM/ODS '99. The structure is
a Ge-Sb-Te recording layer with a ZnS-SiO 2 layer, with an AgO x layer on
top. When illuminated with a laser beam, the AgO x disassociates into Ag and
O, and near-field light is generated along the dielectric interface around the Ag.

A new paper describes the formation of an AgO x layer below the recording
layer that doubles the emission points, boosting the C/N ratio to 40dB (Fig 5),
recording mark length about 200nm. This is about double the level reported at
ISOM/ODS '99.

It was also confirmed that a reduction in the thickness of the ZnS-SiO 2 layer
can improve the C/N ratio for short recording marks, and achieve a C/N ratio
of about 25dB for marks only 100nm long. When the two AgO x layers are
close to each other, the dipole antenna effect comes into play, further
amplifying near-field light.

Amplifying Incident Light 3,500x

Hitachi has proposed an optical head structure that aims to enable areal
recording densities of 180 Gbits/inch 2 or more. It uses a "bow-tie antenna"
design to amplify the electromagnetic field intensity.

The structure and principle of operation are as follows: first, the apexes of
triangular metal patterns are positioned in proximity. When electromagnetic
waves are present, polarizations with differing polarities are generated near
each of the apexes, and polarity reversal is repeated. When conditions for
resonance are met, the structure becomes a generator of electromagnetic
radiation, and waves with sharp intensity peaks are generated through the gap
between the triangles.

The firm has developed a prototype of a triangular metal pattern with a gap of
about 20nm (Fig 6). As is apparent from a simulation, when laser light is
transmitted into the pattern, the optical intensity half-width is a sharp 5nm, and
the peak near-field generated light is about 3,500 times the light that is input. If
this light can be used for optomagnetic recording, the firm believes it will be
possible to achieve an areal recording density of 180 Gbits/inch 2 .

by Masayuki Arai