вход по аккаунту


Materials for Optical Data Storage.

код для вставкиСкачать
Volume 28 . Number 11
November 1989
Pages 1445-1600
International Edition in English
Materials for Optical Data Storage
By Michael Emmelius”, Georg Pawlowski *, and Hansjorg W. Vollmann”
Dedicuted to Professor Wolfgang Hilger on the occasion of his 60th birthday
The outstanding feature of materials research in the eighties has been the convergence of basic
research and practical application, leading to ever shorter cycles of innovation. This is especially true of materials which form the basis of key technologies. The mass storage units of the next
generations of computers will be based on optical processes having a storage density which
exceeds that of all hitherto known storage techniques that are practicable from the technical
standpoint. In view of the fact that since 1982 read-only memories in the form of compact discs
(CD-ROM) have become very successful in the field of audio electronics, research and development are now concentrated on materials for write-once (WORM) and erasable (EDRAW)
storage systems. Suitable materials for optical data storage are substances in which data
markings can be recorded and deleted respectively using semiconductor lasers. Materials
development is centered on the synthesis of infrared-absorbing dyes for WORM memories and
the production of rare earth/transition metal alloys for magneto-optical data recording. An
introduction to CD-ROM technology will be followed by an overview of the state of development of the most important storage materials which are currently available commercially, and
then the properties of these materials will be discussed with reference to selected examples.
1. Introduction
The storage of biological information represents one of
the basic explanations for the existence of highly differentiated living organisms. The biological information medium
D N A (deoxyribonucleic acid) is the substrate for numerous
Dr. M. Emmelius, Prof. Dr. H. W. Vollmann
Hoechst Aktiengesellschaft
GeschXtshereich Informationstechnik
Werk Kalle/Alhert
D-6200 Wiesbaden (FRG)
Dr. G. Pawlowski
Hoechst Aktiengesellschaft
Zentralforschung I
D-6230 Frankfurt/Main 80 (FRG)
I n f . Ed Engl.
28 (1989) 1445- 1471
enzymes which, at the molecular level, store, read and erase
genetic data. Combinations of the basic building blocks of
D N A (uracil, adenine, cytosine and guanine) provide the
code for protein amino-acids as well as addresses on the
D N A for starting and stopping the D N A replication. Replication, which is equivalent to storage, commences at primerformatted D N A segments. Because of corrective mechanisms, for appoximately every lo1’ nucleotides combined, a
maximum of one unit is incorrectly positioned. Erasure of
genetic information occurs either during regulation of cell
metabolism o r as a result of chemical mutations of the base
pairs. On considering the interrelationships of molecular biology from this rather formal point of view, parallels with
the function of technical data stores are immediately evident.
At the same time it is clear that what is involved here is a
~ ~ r l u f i s f i i ~ . ~ ~ ~ml lh. ~
Hc, lD-6940
i u ~ f WCJinhrim.1989
0570-0833i8911/ I 1.1445
X 02.5010
fundamental principle,''] which is implemented in nature
and to which the physical structuring of information systems
can be traced.'']
Not only d o biological and technical information carriers
differ in numerous fundamental respects; they differ, above
all, as regards the physical size and arrangement of their data
units. The genetic information of a DNA molecule corresponds to an information density of lo2' bit cm-3 (three-dimensional storage), while technical memories of the highest
information density hold approximately lo8 bit cm -'(twodimensional storage) (Fig. 1). In spite of the high rate at
which miniaturization is progressing, as well as the use of
new storage processes, it appears dubious whether the limit
to storage at a molecular level can be overcome in the near
future.[*' However, optical memories represent an important
step in this direction, since they exceed the storage densities
of hitherto well-established information media.
Research has been in progress on optical memories since
the end of the sixties. However, it was not until 1982 that the
breakthrough finally came with the audio compact disc; this
was closely linked with the availability of new semiconductor
lasers as light sources. In the field of consumer electronics,
See also the Editorial Essay by D.Hoarer in the AdvuncedMareriak part of
this issue.
i o 7 f
I 980
Fig. 1. Trend toward higher storage density for various media. Reduction of
area A required to store one bit of information (after [2c]).
disk file. A
magnetic bubble, 7 thin film memory. 0 optical disk, 0 semiconductor.
the conventional playback processes lost considerable
ground, since it was not possible to achieve comparable levels of quality of reproduction using any other medium. Accordingly, it was an obvious step to utilize the potential of
Michael Emmelius was born in 1954 and studied chemistry and medicine at the University oj
Mainz, where he was awarded his doctorate for studies in the field of macromolecular chemistry
under Professor H. Ringsdorf. In 1986 he joined the research department of the Information
Technology Division at Hoechst AG and was entrusted with the development of photoresists. In
1988 he became assistant to the director of R&D in injormation technology. Since September
1989 he has been concerned with technical marketing. His scientific interest pivots around the
physicochemical problems of materials research.
Georg Pawlowski was born in 1952 and studied chemistry at the University of Tiihingen where
he completed his doctoral work on conductive polymers under Professor M . Hanack. In 1981 he
joined the research department ofthe Information Technology Division at Hoechst AG where his
work was primarily concerned with "photoreactive polymers". In 1987 he became assistant to the
director of R&D in information technology. He is at present leader of the photochemistry group
in the central research department at Hoechst AG. His interest centers on the development of new
radiation-sensitive materials.
Hansjorg K Vollmann u'as born in 1933 and studied chemistry in Cologne, Bonn and Stuttgart.
After gaining his doctorate $or studies in preparative organic chemistry under Professor H.
Brederek he workedfor a short while as research chemist at Cussella AG. In 1964 he joined the
central research department at Hoechst AG and worked in the area of intermediates and
dyestgfls. In 1970 he was appointed in charge of the construction and direction of the central
research department of the American Hoechst Corp. (now Hoechst Celanese Corp.) in Rhode
Island, USA. Since 1978 he has been head of research and development in the Informalion
Technology division at Hoechst AG. He is honorary projbssor .for Industrial chemistry at the
University of Wiirzburg. His main scientific interest fbcuses on the products, processes and
applications of photochemistry in information technology.
A n g e n . Chrm. I n f . Ed. En,$
2X (1989) 1445-1471
optical memories for the greatly expanding computer technology, since it is mainly the requirements of modern microelectronics which stimulate the search for faster and more
powerful mass storage units. Circuits exhibiting very large
scale integration permit access to 4MByte DRAM (Dynamic
Random Access Memory) from semiconductor memories
and to the management of sophisticated software, which
require permanent storage of increasing quantities of data.
At the same time, there is a trend towards information storage in a more compact and robust, i.e. user-friendly, form.
Accordingly, recent years have seen a broadly based advance
in the development of appropriate materials, which has led
to the initial marketing of optical data memories.
The present paper is concerned with the development of
materials for optical data storage and gives special consideration to systems which can be applied on a technical scale
and which are of commercial interest. An introduction to the
most important concepts and processes will be followed by a
discussion of the principles of Compact Disc and CD-ROM
(Compact Disc Read Only Memory) technology. The broad
principles of optical readout and of the production of polymer storage discs with impressed information will be summarized in Section 3. Sections 4 and 5 give an overview of the
most important materials and their properties for write-once
(Write Once Read Many, WORM) and erasable (Erasable
Direct Read After Write, EDRAW) storage media. The main
items covered are organic dyes for WORM systems and alloys of rare earth elements and transition metals for
EDRAW systems.
There will be no discussion of metal coatings for WORM
memories. More extensive references deal with compact disc
and CD-ROM memories" - 5 * 'I and with general aspects of
optical memorie~,'~']
applications for optical data stores [*I
and recording media for WORM and EDRAW memories,[6.9- 181
reflectivity of the recording medium is altered by the marking process. This takes place locally at positions heated by
the laser beam so that data spots are created, which are
distinguished from the surrounding medium by their reflectivity. The physical marking process of EDRAW media is
reversible, and thus enables the user to erase data which have
been inscribed. EDRAW memories have been developed on
the basis of magneto-optical materials, reversible phase
transformations and photochromic dyes. In the first case,
use is made of the magneto-optical effect: the rotation of the
plane of polarization on reflection or transmission of laser
light on or through magnetized material, in order to detect
magnetic domains as data spots. The reversible phase change
between crystalline and amorphous phases leads to an alteration of the reflectivity of the storage medium. The photochromic effect permits the marking of data on the basis of
the differing absorption of a tautomeric dye molecule. Thus,
in contrast to the two previously described storage methods,
photochromic systems require the use of two laser sources.
In addition, there exists a series of other reversible storage
processes to which no technical significance is yet, however,
The development of optical data storage is closely connected with the progress made in microelectronics. This applies both to the production of semiconductor lasers and also
to the entire process technology which is necessary in order
to produce structures in the pm range. The photoresists used
for this purpose were primarily developed for the semiconductor industry, as were numerous vacuum coating processes for the preparation of metallic films. Accordingly, it was
possible to use the know-how in vacuum technology for the
construction of appropriate production plants for optical
storage disks (Fig. 2).
2. Optical Memories
Audio and video CDs are optical memories which are
based on a principle similar to that of gramophone records:
information items are encoded in a structural pattern, from
which the original information can be regained by means of
a playback unit. In contrast to gramophone records, the
readout takes place by a noncontact method using laser
light. The information is impressed into a plastic plate and
cannot be altered. CDs have meanwhile been used for other
fields of application as well, outside the field of consumer
electronics. The term CD-ROM has established widespread
usage in connection with this type of information storage.
This term signifies Compact Disk Read Only Memory (Data
disk with k, in contrast to the audio/video disc with c). Writeonce optical memories are designated as WORM systems. In
contrast to the CD-ROM, it is possible for the user to record
information. The principle of all WORM memories is based
on an irreversible physical transformation of the recording
medium, which is initiated by the irradiation of laser light.
The following processes have been described?" formation of
pits, bubbles, bumps, changes to the surface texture, phase
change, switching of liquid-crystal phases, alloying of metal
films and photochromic color change. In most cases, the
Anpcu.. Clitwi. I n ( . E d EngI. 28 ( 1 Y N 9 ) 1445- 1471
Fig. 2. Vacuum coating machine for the production of magneto-optic disks. To
reach the required clean room conditions the room is air conditioned with
laminar flowing air, which is fed in through a filter in the roof and o u t through
a floor grating.
The problems which are connected with particle contamination and its effect on production output are common to
both the production of optical storage disks and the semiconductor industry. Both cases require processes with ex1447
tremely stringent purity demands and acceptable particulate
concentration tolerances.
3. Compact Disc Technology (CD-ROM)
Not only has CD-ROM technology laid the foundations,
in terms of process technology, for optical information storage it has in addition stimulated the development of other
storage systems (WORM, EDRAW) which are advanced
and therefore in many ways significant for the future. The
key to all optical processes resides in the availability of suitable radiation sources which, for mass production, must
above all be small and cheap. With this background, it was
the progress made in laser technology which, together with
the development of semiconductor lasers, fulfilled all prerequisites for the successful marketing of optical storage systems. The experience gained from microelectronics made it
possible to achieve miniaturization and the availability of a
process technology for lasers which are suitable in terms of
quality and price. A point of prime importance in the technology for the production of compact discs is the multiplication of the original information, in a manner similar to that
used in the stamping of gramophone records. In the case of
the CD, however, the size of the impressed structureswhich is 0.5 to 2 pm-is several orders of magnitude smaller
than those of a gramophone record. In order to produce
corresponding dies and to use them in special injection molding processes, it was therefore necessary to find appropriate
techniques which could be developed so as to include methods and materials known in the semiconductor industry. The
outstanding quality of reproduction, which is highly important to the purchasers, and the decreasing cost have helped
this new technology to achieve great market successes, especially in the audio sector.
This is not yet true of the CD-ROM data disk, which can
store up to 650 Mbytes of information. A whole series of
data banks are supplied in the form of CD-ROM disks, but
in many cases there is no compatibility with the current
operating systems and the standardization of interfaces and
peripherals is proceeding only slowly. In spite of significant
cost advantages (Fig. 3 ) , certain technical problems such as,
for example, the long access time of approximately 0.5 to 1 s,
are still preventing widespread acceptance on the market.
Philips and Sony have initiated a joint project for the
purpose of increasing the attractiveness of the data CD, and
in 1987 specified the CD-I (Compact-Disk Interactive Media) in the so-called “Green Book” (forum for publication of
standards). This amounts to a closed system which, using the
multitasking operating system 0s-9, is designed for microprocessors of the 68 family. As yet, there is no compatibility
with MS-DOS. A further step in this direction is represented
by the “Extended Architecture” which was first designed by
Philips, Sony and Microsoft in August 1988 (CD-ROM
XA). Using this system, it is also possible to use the data CD
to store digital audio data and graphics in a similar manner
as the CD-I is used but on a system-independent basis. While
it is probable that new playback systems will be necessary for
the CD-ROM XA, a different format can be operated with
the CD players previously available: DVI (Digital Video Znteractive), which is based on a 100: 1 data compression and
decompression for digital video and should facilitate the PCassisted processing of images. The beginnings of the technology of optical data storage go back as far as the late fifties;
marketing in the form of the audio compact disc did not start
until after 1982, as a result of the collaboration of Philips
(basic research and disk development) and Sony (optics and
playback unit). Following the immense market success of the
new audio technology, several companies have become established as CD manufacturers, and joint efforts are meanwhile being made in the development of this technology by
large companies: in 1985, Philips and DuPont founded PDO
(Philips DuPont Optical), a joint venture company which is
concerned with research and development of optical storage
disks and which operates with Polygram (Hannover, FRG),
the largest CD production site in the world.
The progress made in laser technology was tremendously
important for the development of optical data storage. The
first CD players were, in some cases, still operated using
He-Ne or argon ion lasers. Since that time, all players have
been equipped with semiconductor lasers, which can be produced inexpensively and in a more compact form. The widespread technical use of semiconductor lasers (emission wavelength ca. 780-840 nm) became possible only with their
continuous operation at room temperature and long service
To this end, it was necessary to reduce the threshold
current of the lasers: the electrical power which is expended
until the laser threshold is reached is lost in the form of heat
and rapidly leads to the destruction of the optical system.
The availability of appropriate semiconductor materials and
the progress which has been made in the field of production
engineering have created a situation in which semiconductor
lasers can nowadays be supplied at very favorable prices (ca.
US $ 3 each).
3.1. Digital Data Storage and Reproduction
Fig. 3. Cost (US $) per bit and access times of classical and modern storage
media pel.
Audio CDs and video CDs contain data in digitalized
form which are read by means of a playback system (CD
player consisting of disk drive and optical system) and are
transmitted via analog-digital conversion to the output unit
(loudspeaker and/or screen). The information carriers are
pits which are impressed into the surface of a plastic disk and
are separated by the relief positions, the land (Fig. 4a). The
pits, which are arranged in sequence and which have a depth
of approximately 0.5 pm and a width of 0.8 pm, form tracks
at a spacing of approximately 1.6 pm.13,41
The disk consists
of an injection-molded polycarbonate or polymethyl
methacrylate substrate (diameter 12 cm) and is coated on the
Fig. 4. Schematic representation of the construction of optical disks. a) CDROM disk. b) WORM disk as air-sandwich construction with a tellurium alloy
as recording layer. c) EDRAW disk as three layer construction with a magnetooptic terbium-iron alloy as storage layer.
Angew. Cham. h i . Ed. Enxl. 28 (IY89) 1 4 4 S I 4 7 1
information-carrier side with a highly reflective aluminum
film, which is protected against mechanical damage by a
polymer layer.
The playback unit includes an optical system which is
located on a pivotable arm and is tracked to follow the pit
track. The optical system focuses the light beam of the laser
through the substrate onto the plane of the pits; the beam
has a diameter of approximately 1 pm on the disk. The light
reflected by the aluminum mirror is passed to a photodetector. When a pit is swept over, the signal intensity of the laser
beam decreases significantly since the light is scattered by the
pit structures. The optimal reflection conditions are on the
land situated between the pits. In this way, it is possible to
read binary data which are encoded in accordance with specified algorithms: reflection results in a high signal intensity
and is recorded for example as 1; a signal intensity which is
reduced by scattering represents 0.
The binary data form of the analog original sound recording is obtained by a frequency-dependent recording in audio
channels: the signal level of each channel is given as a binary
code word of 16 bits, to which error correction processing
bits are also added. This gives rise to a bit stream encoding
the sound recording. On playback of a CD disk, the decoding of the square-wave signal takes place in the electronic
system of the recorder. Using electronic filters, the bit stream
is broken down into the individual signals for sound (audio
CD) and also for picture and color (video CD). In the case
of a video CD, the picture is encoded by means of the spatial
frequency of the pits along the track, while color and sound
are encoded by modulation of the ratio of pit length to repetition rate. The signal fulfills additional functions which are
essential for the recorder to operate properly. Disk rotation
speed as well as focusing and tracking of the laser beam must
be measured continuously and, where appropriate, automatically readjusted.r20]These servo functions avoid situations
in which the laser beam, for example, “deviates” on the disk,
jumps between pit tracks or “overlooks” pits. By means of
special photodetectors, consisting of four subquadrants and
each delivering a respective signal, several signals are generated electronically from the intensityitime progression of a
laser beam, and these signals are available for servo functions. Another possibility consists in three beams being produced optically from a laser source, and then being used to
scan the disk surface. These three additional signals are thus
available for adjusting the optical system and the disk. Finally, it is necessary to constantly regulate the intensity of the
laser light in order to maintain the optimal signal-to-noise
ratio, and thus to obtain the same conditions for reading the
information when the disk is played back at any position.
The compact disc system developed by Philips and Sony
has meanwhile become a world standard, and has been licensed to many other companies. There are as yet no standardized formats for the WORM and EDRAW data disks.
This applies both to the disk construction and to the write/
read unit. One construction for the WORM data disk and
one for the EDRAW data disk are shown in Figures 4 b and
4c respectively. These represent some of the numerous systems which are currently at the development stage.
As far as the data disks are concerned, the main difference
from the CD format is that the substrates are provided with
one or more guide tracks which, using a laser beam, permit
orientation on the initially uninscribed disk. The impressing
of the guide track, also referred to as a groove or track,
creates elevated regions to receive the data markings. A series of possible formats are distinguished according to the
mode of rotation of the disk and the form of the guide
track:r3.41CLV (Constant Linear Velocity), CAV (Constant
Angular Velocity), spiral and concentric track arrangement.
Where the speed of rotation of the disk is constant (CAV),
the disk has linear speeds which are different inside and
outside, and the data markings are larger outside than inside.
Constant pit size requires constant linear velocity, and in this
case the speed of rotation must be regulated as a function of
the radius. For orientation purposes, marks are used within
the guide track system, which divide the disk into sectors. If
the marks are set at constant intervals beside a spiral track,
this is referred to as the “sampled servo” format. In the
continuous composite format tracking is achieved by direct
evaluation of the signal reflected from the tracks. Due to the
division of the CDs into sectors, the data file can be stored
in a similar way to that applicable with the magnetic disk
memories, not only on a serial basis but also in a dispersed
configuration (sector offset with interleave factors). The format of a disk controls the maximum attainable storage density, so that the storage density is increased by appropriate
changes to the formatting. For this purpose, it is, e.g., possible to write data directly by using track grooves. The secondorder reflected beam then reproduces the information from
only one side of the groove.’”]
The air-sandwich design (Philips) is shown in Figure 4 b122a1as an example of the design of a write-once data
disk. The recording layer, a tellurium alloy, is coated onto
the substrate which is provided with track grooves. Using
appropriately dimensioned spacers at the center of the disk
and at the edge of the disk, the size of the air gap is regulated
when two disks are welded together. The composition of the
gas in the gap can be chosen so that the air-sensitive layers
are protected. However, it is still possible for the recording
layer to be subjected to attack as a consequence of failures in
the sealing of the disk construction; accordingly, in another
design the air gap is filled with an inert polymer mixture.122b1
In addition to the direct coating of the substrate disk with a
light-sensitive write layer, it is also possible to apply the
information layer to a flexible fabric which is firmly clamped
in a suitable disk construction (Sealed Disk Assembly, SDA;
Kodak).[22‘1 When data disks rotate (approximately
1800 rpm), in some cases relatively large forces act, and these
have the effect of bending the flat substrate (radial run out);
this can lead to difficulties in focusing and tracking the laser
beam (cross talk). Thus, the suspension of a light-sensitive
lightweight layer achieves the greatest possible mechanical
decoupling between disk and layer.
The construction of magneto-optical storage syst e m ~ (EDRAW,
~ ~ ~ see
, ~Fig.
~ 4c)
~ is similar to that of
WORM disks. A recording film made from rare earth elements and transition metals is located on a substrate provided with track grooves, between two protective layers consisting of amorphous silicon nitride o r aluminum nitride.
Encapsulation is necessary for the purpose of protecting the
storage medium from oxidation. A polymer adhesive is applied to the outer protective layer in order to produce, by
combination with a further substrate. a disk which can be
inscribed on both sides. In addition to the protective function, amorphous nitride films also have the function of effecting the coupling-in and coupling-out of the laser light for
optical writing, reading and erasure.
In the case of magneto-optical memories, two disk formats
are currently under development: 3.5” (approximately
130 MByte) and 5.25” disks (approximately 300 MByte).
The two formats will probably be used for different markets.
3.5” disks and disk drives could be used for personal computers. The optical storage disk can fulfill two functions at the
same time which previously would still have required separate components: the hard disk function as mass memory
and the floppy disk function as exchangeable data file. The
5.25“ system will mainly be used for data archiving and data
exchange at very high data rates.
In the case of the optical WORM and EDRAW systems,
the operations of data encoding and decoding correspond to
a Iarge extent to the principles applied in magnet technology.
Accordingly, in the field of computer operating systems it is
necessary to make changes on introducing the optical storage technique only insofar as they relate to the matching of
the interfaces to the optical disk drive. All data disk systems
which are currently exhibited are to a large extent compatible
with existing systems.
3.2. Production of Substrates
One of the prerequisites for the acceptance of new massproduced products is high quality production, as well as high
output in short periods of time a t the lowest prices possible.
Accordingly, the development of C D disk production was
pursued, from the outset, with particular regard to these
conditions, and has made a contribution to the success and
commercial attractiveness of the C D technology.
There are essentially three processes for the multiple production of flat molded components: stamping, injection
molding and injection stamping. The first represents a process which is widely used in the gramophone record industry,
but which is not suitable for use in C D production since the
requirements as regards dimensions and mold tolerances are
more stringent by an order of magnitude (the submicron
range). Injection molding and injection stamping can be used
to achieve the desired structural interrelationships, injection
stamping representing the most demanding technique. Both
processes are used in the production of CD audio and video
discs and in the production of substrates for WORM and
EDRAW disks. The production is divided into the fabrication of the master and the preparation of substrates. Injection-molding dies are cast in a multistage process (mastering)[z83301 which involves the latest technologies of microelectronics (Fig. 5). The mastering process begins with the
cleaning and polishing of a glass disk which is required to
construct the master. An adhering layer is applied by vapor
deposition using an agent creating water-repellent properties, e.g. hexamethyldisilazane (HMDS), which coats the surface with a positive photoresist. The surface is exposed to a
UV argon ion laser, which inscribes the pit structure in accordance with the digital information as played back from
the master magnetic tape.[251During subsequent development the exposed resist is dissolved away. At this stage it is
Angeu.. Chem. Int. Ed Engl. 28 (1989) 1445-1471
gence. Bubbles, gel particles, inclusions, surface defects, and
zones of turbidity due to impurities can give rise to scattering
and absorption effects which cannot be compensated by the
servo functions of the electronic and optical systems. Significant properties of the disk materials with regard to mechanics, optics, environment and productivity are represented in
Figure 6 for polycarbonate (PC), polymethyl methacrylate
(PMMA), polycarbonate-polystyrene blend (PC/PST) and
Fig. 5. Mastering process of the stamper for the injection molding optical disk.
necessary to precisely control the development process since
the geometry of the pits is influenced unfavorably by underdevelopment or over-development; this also applies, in the
same way, to the imaging of the track guide grooves in
WORM and EDRAW mastering.[291The photoresist and the
glass surface are then coated with a layer of metal (nickel in
most cases) approximately 50-100 nm, either by vacuum
vapor deposition or by sputtering. In this way, the master
surface can be visually monitored before further processing
and can be directly played back in a disk drive; an initial
quality control is possible by comparing master and the master magnetic tape.
Further metallic layering to form the actual metal master
is carried out using a currentless or electrolytic
After separating the master from the glass base, parent masters are produced by electroplating, and the individual injection-molding masters are formed from these parent masters
by means of a second electroplating operation. By setting up
such a “family of masters”, it is possible to prepare new ones
immediately without repeating the costly mastering process.[261The mastering process specifies the quality of the
finished disk, and accordingly numerous attempts have been
made to improve critical process
Within narrow tolerances, the substrate materials for optical data storage systems must exhibit properties which are
mainly determined by the nature of the data storage and
reproduction. In most cases, the information is read through
the disk, i.e. the laser beam must pass through the substrate
material without loss of intensity and-in the particular case
of magneto-optical systems-without change of polarization. This requires a substrate material of high optical homogeneity, purity and optical transmission and low birefrinAnpeir. Chein. Inf. Ed. En,qI. 28 (1989) 1445-1471
Fig. 6. Properties of polymeric substrate materials [32]. Parameters are represented regarding optical, mechanical and environmental properties and productivity of polycarbonate, polymethyl methacrylate, polycarbondte-polystyrene
blend and modified polymethyl methacrylate. The performance of a polymer
improves from the inner to the outer circle.
modified polymethyl methacrylate (Mod. PMMA).13*] A
further criterion for the selection of a substrate material is its
absorption of water, since swelling and thus distortion of the
disks can arise as a result of water vapor. The sensitivity to
moisture of certain recording media is very great, so that the
water absorption of the substrate may be the limiting factor
with respect to its
The severity of this problem may
be reduced by the use of protective layers; however, it can
represent a latent uncertainty regarding the long-term stability of data archives. It is possible to measure such effects
quite well by accelerated storage tests; the manufacturers
give their guarantees on the basis of these tests (required
long-term stability: 10 years).
The birefringence of the flat polymer disks must be as
small as possible; for example, it is specified at f50 nm for
the Compact Disc. Birefringence occurs where a light wave
in a medium propagates at differing velocity.[36.371 Prerequisite for this is an anisotropic polarizability of the medium. In
the case of a polycarbonate, this preferred direction of the
polarizability is parallel to the polymer chain. Accordingly,
the resultant birefringence is referred to as orientation birefringence. When forces act on a molded component of polycarbonate the birefringence alters in consequence of the
movement of chain segments. In the course of the production
Fig. 7. Birefringence of injection molded polycarbonate substrates. The left
disk fulfills the specification for an optical disk, the right disk is useless because
of inhomogeneities.
process, for example by injection molding, it is possible for
deformations to be frozen in as internal stresses. The resulting stress birefringence remains operative at room temperature; it may decrease or entirely disappear only at higher
temperatures and on reaching the glass point.[341The process
of molding storage disks must accordingly be adapted to the
specific birefringence properties of the polymer; in addition
to this, it is necessary to include the effects of other production steps. The uniformity of the birefringence of disks can
easily be monitored visually, for example, by placing substrates between two polarizing foils (Fig. 7). Reduction of
the birefringence of substrate disks may be achieved by the
use of polymer blends which consist of polymers having different anisotropic polarizabilities. Thus polycarbonate/polystyrene,[391polyphenylene ether/poly~tyrene,[~’’
methacrylate/polyvinylidene fluoride[381 and polymethyl
methacrylate/polyvinyl chloride[391blends can be prepared
which are suitable for the production of disks. New types of
polymers such as cyclic polyolefins, compounds with very
low water absorption, are now being used for the production
of substrates.
Instead of using injection molding or injection stamping,
it is also possible to produce CD disks by a process developed by Philips (Fig. 8).r30,40-431
In this “2P Process” (pho-
Fig. 8. 2P process for disc replication (2P = photopolymerization). [30] a) Production of plate and stamper. b) Filling with 2P polymer. c) Photopolymerisation. d) Separation of disk from stamper.
topolymerization) a base plate is fixed at a defined spacing
on a nickel stamper and the gap is filled with a photopolymerizable mixture. The mass is hardened by means of UV
radiation and removed from the stamper together with the
base plate as a completely structured disk. An advantageous
feature of this process is the precise structural imaging, since
the Curable mixture can flow into every fine pit of the stamper. This process may be carried out using simple devices,
and the quality of the raw liquid material is controlled at low
costs. Moreover, the wear and tear of the stamper in this
pressureless technique is considerably less than in other processes, whereby additional costs can be saved. The photopolymerizable mixture is composed, for example, of ethylene
acrylates o r ethylene methacrylates and photo initiator^.^^'^
The problems associated with this process reside in long
exposure times involved in the photopolymerization. The
cycle times of injection molding machines are in the region of
five seconds for the molding of a disk, but the UV curing of
a mixture demands considerable optimization to achieve exposure times of similar brevity. Monomers and photoinitiators must virtually be 100% consumed in order to eliminate
subsequent complications by diffusion into the storage medium. In the course of polymerization, a decrease in volume
usually takes place as a result of the combination of the
monomers, and the molded component shrinks accordingly.
In the case of the 2P process a considerable proportion of the
process-technology know-how is specifically based on keeping the volume shrinkage to a minimum.
4. Write-Once Storage Disks
The development work carried out on write-once optical
storage disks has led to the first commercially available products which are marketed in combination with appropriate
write/read systems. Write-once optical storage media are
designated as WORM or DRAW (Direct Read After Write)
systems; common to these is that they accept information
from the user and can make this available again within short
access times. WORM and DRAW systems differ from each
other in that the latter can compare the stored with the
specified information immediately after each inscription
process, and thus correct readability is guaranteed.
A typical field of application for WORM memories is the
archiving of data. Their objective is thus to replace the documentary or microfilm archiving which is customary nowadays in offices. With a diameter of 5.25 inches, commercially
available storage disks offer a storage capacity of approximately 2 x 400 MBytes, which corresponds to the information content of approximately 150 000 typed A4 pages. This
mode of office filing, which is still unusual, will create new
possibilities for information processing in view of the increase in direct electronic communication via telefax and
electronic mail. Furthermore, write-once optical storage
disks, just like CD-ROMs, may be used as data banks for
reference works. In this case, the mixing of pictures, texts and
data is possible. In addition, compact and fast-access
“analog” image storage in the video field is also possible with
these storage systems.r441
While the advantages of write-once optical storage disks
as compared with conventional filing (long access time. great
A n g m C‘hem. In1 Ed. EngI. 28 (1989) f445 -1471
bulkiness, labor-intensive), microfilming (long access time,
no freely selectable sorting facility, costly setting-up procedure) and storage by means of magnetic tapes (medium access time, inadequate data security, limited data service life)
are clearly evident, they find in the magnetic disk systems,
which are likewise under rapid development and which have
exceptionally short access times, a genuine competition over
which they have, to date, not yet been able to acquire precedence.
In order to compete against the magnetic storage technology, the following fundamental requirements must be placed
on write-once optical storage disks: high storage capacity
(> lo8 bit/cmz), fast access times (< lo-’ s), low storage
costs ( < l o - ? DM/bit), high data service life (> 10 years),
high write sensitivity ( < 1 nJ/bit), reliable reproduction as a
result of a high signal-to-noise ratio ( > 55 dB), low defect
density ( < 10- bit after correction) and easy handling of
the entire system.
4.1. Principles of the Write/Read Mechanism
The selective, thermally induced and irreversible deformation. as well as the phase change of an active layer, the
so-called recording layer, have become established as preferred methods for the inscription of write-once optical storage disks. In practice, radiation from semiconductor lasers
having emission maxima between 780 and 840 nm is used.
This absorption of energy results in a locally specific chemical-physical change of the active layer. As a result of this
change, which is limited to a very small area and which is
unambiguously defined, it is possible by means of a suitable
reading mechanism to detect digital information and to convert it into a bit sequence.
The most widespread technique is based o n a n apertureforming mechanism (ablative recording) (Fig. 9 a). The ac-
Fig. 9. Conventional layer configuration for ablative recording
tive layer converts the closely concentrated irradiated laser
light into thermal energy, which at approximately 1700 “C
initiates complex physical processes.[451 This leads to the
formation of a pit with the simultaneous localized alteration
of the optical properties of the layer which are detected by
means of the intensity variation of a weak laser beam. In
order to convert the laser light into heat, the recording layer
AngiJw. Chcn? Int. Ed Engl. ?X (1989) 1445-1471
must consist of a material which absorbs radiation of the
prescribed wavelength. Possible materials used are metals or
amorphous dye films.
The basic structure of this type of write-once storage disks
consists of a disk-shaped molded substrate unit, which is in
most cases solid in structure and usually made of glass, aluminum, polymethyl methacrylate or polycarbonate. The
recording layer is applied to this substrate unit and covered
with a protective layer. Occasionally, an additional layer is
introduced between substrate and recording medium; this
improves the adhesion properties, reduces the microporosity
of the substrate, and, where appropriate, prevents the diffusion of components of the active layer into the substrate. The
microstructure of the substrate has a substantial influence on
the background noise and thus on the quality of reproduction of a disk, which is characterized by the signal-to-noise
ratio (SNR).
The active layer consists of a metallic o r dye film, which
has a thickness of approximately 50-200 A and which, depending upon the material, is applied by sputtering, sublimation, o r spin-coating from a solvent mixture, which is financially preferred but is only possible with soluble dye syst e m ~ . [ The
~ ~ ]long-term stability and optical constancy of
pure dye films[471is frequently limited as a result of agglomerate formation and the tendency to crystallize. These effects
can be circumvented by using mixtures of the dyes with
special polymers,[481which not only considerably improve
the stability of the recording layer but also the quality of the
pit formation (dye-in-polymer (DIP)
In addition to these, recording layers consisting of special polymer
composites are also used. 1501Gupta modified the DIP configuration by applying a thin transparent ceramic layer to the
recording layer[511(Fig. 9b). In the course of the writing
process, a pit is formed in the active layer and gas is evolved.
The ceramic layer is likewise heated and bows as a result of
the internal pressure generated in the pit. The bubble formed
above the pit leads to a unification of the pit geometry and
the still intact ceramic layer prevents thermally induced ejection of material. Both effects result in an outstanding signalto-noise ratio of the medium ( >65 dB).
Another variant consists in a separation of the functions
of the recording layer into an absorption layer and a film
forming pits (Fig. 9c).L5’] This arrangement permits a fine
optimization of the respective layer with regard to its specific
task. A disadvantage of these multi-layer variants is that the
additional layers lead to an increase in the layer defects and
thus an increased bit error rate (BER) and rejection rate.
The LIDA concept (Laser Induced Dye
is based on the diffusion-brought about by laser light and
thermally induced-of a dye into the substrate polymer
(Fig. 9d). The resulting difference in the optical properties
between pure dye film and molecularly dispersed dye is used
for signal differer1tiati0n.I~~’
Alternatively, it is possible to apply a temperature-sensitive layer beneath a metallic absorbing layer which is irreversibly expanded by the irradiation, for example by the
evolution of a gas, and thus leads to bubble formationf551
(Fig. 9e). The altered light scattering of the reading beam is
produced by the deformation of the film surface. A writeonce storage disk using this arrangement has been exhibited
by Thomson C S F under the designation of Gigadisk.
The formation of pits or bubbles has the effect of altering
various optical properties of the recording layer, e.g. extinction, transmission, reflection or light scattering. In most cases, the readout of the information takes place by the write
laser operating with relatively low radiative power, o r occasionally by a read laser emitting a t a different wavelength. In
order to be able to inscribe and to read an optical storage
disk on both sides, use is made of the altered reflection behavior of the active layer. The signal contrast can be considerably improved by the use of an additional thin, metallic
reflective layer beneath the active layer. The mechanisms of
pit and bubble formation in the active layer have been intensively investigated; the initial studies date back to
Mayden[561and Kivits et al.'571, who used thin tellurium
films. According to Kivits et a1.i581,the pit formation takes
place in several stages. When the laser pulse strikes the metal
foil, optical energy is converted into thermal energy; and
within the narrowly defined irradiation region, temperatures
are reached which are close to the boiling point of the metal.
The duration of the absorption of energy during the laser
pulse is so short that it is not possible for any thermal equilibria to become operative (superheating). The temperature
profile occurring in the molten metal leads to surface tension
gradients (Marangoni effect),[45.49b1 as well as inhomogeneous pressure, bubble formation, and vaporization zones
which lead to flow of the material and thereafter to a local
broadening of the active layer with pit formation. As a result
of the rapid cooling of the material after termination of the
laser pulse this condition is frozen in, and the pit and edge
geometry is determined. Comparable interrelationships may
also be utilized regarding the pit formation in active layers
based on organic
The dye molecules
which undergo electron excitation upon irradiation convert
the absorbed energy into vibrational, translational and rotational energy, resulting in thermal energy. On account of the
low thermal conductivity of organic compounds, the heating
is limited to a narrow local region. This heating leads to
melting and vaporization o r even decomposition of the
molecule. The pit formation in organic materials is discussed
controversially and is associated with thermally induced surface tension and melt viscosity phenomena,["l the optical
density of the recording layer1601or the rheological properties of the active
In this case, it is necessary to find a compromise between
film thickness and dye loading on the one hand and pit
formation on the other, since the growth increases the absorption capacity, but in order to form pits, more material
has to be discarded.
As compared with metal films, recording layers containing
organic dyes which absorb IR radiation have numerous advantages, such as an increased resistance to atmospheric influences and lower toxicity.['3, 's] In addition to this, especially in the DIP configuration, they combine high
physicochemical stability, low thermal conductivity and relatively low melting o r sublimation temperatures.
4.2. Dye Systems and Dye/Polymer Systems
The traditional task of dyes consisted in generating the
coloration of requisites found in daily life. Consequently, the
efforts put into their synthesis in the past were in particular
directed towards providing dyes having absorption maxima
in the visible range, i.e. between 420 and 700 nm, while dyes
for absorption in the near infrared (NIR range) became of
technical interest only with the advent of infrared photography. Accordingly, it is no surprise that virtually all dyes
manufactured on a large industrial scale are of no interest for
the purposes of optical data storage, since they exhibit no or
only slight absorptions in the NIR range, while these absorptions are necessary when using diode lasers to achieve the pit
formation temperatures in the write process and signal contrast levels in the readout process. However, considerable
progress has been made in the meantime specifically in the
synthesis of NIR-sensitive dyes.
If dyes are used as absorbers in write-once disks, then the
following general requirements must be met: high absorption capacity for semiconductor laser radiation (780 840 nm), in order to require low write energies; a commensurate threshold value, which permits a distinction between the
write and read process at the same wavelength; low thermal
conductivity in order to achieve pits of sharply defined
shape; good film formation; high stability in relation to
light, heat and moisture; low toxicity; high thermal stability
(for sublimates).
Depending upon their structure, dyes may be applied to a
substrate by sublimation o r spin-coating. Thus, without reference to the material concerned, it may be stated that according to the hitherto published results systems which have
been applied by sublimation are superior to spin-coated dye
solutions as regards stability towards atmospheric effects
and the quality of pit formation, but are inferior in terms of
signal-to-noise ratio. With frequent reading, the latter is
caused by sublimation and leads, as a result of deposits, to a
surface roughness which reduces the uniformity of the reflectivity of the layer.
4.3. Dyes for WORM Memories
The initial investigations on write-once disks based on
dyes were carried out using non-directly-modulable gas
lasers, for example argon ion lasers emitting at 488 nm,
owing to the lack of NIR-sensitive dyes. Even at that time,
the benefit of this technology became known and was protected under patent law. The dyes which were used included
azo dyes,1621k e t o c o ~ m a r i n s , q~ ~ i~n~o] n e s , [ ~fluoresce~~I
in,f631polyester y e l l o ~ , [ ~ 601
~ , ~o r~d"i ,e t h o ~ y t h i o i n d i g o . ~ ~ ~ ~
P h t h a l ~ c y a n i n e s , ' ~naphthoquinones,[661
simple squaryliurnr6'] and cyanine dyesf681 o r triarylmethane dyesf691
proved to be suitable for recording information by means of
helium-neon lasers (633 nm). The practical significance of
these storage media is slight. Nevertheless, it should not be
overlooked that it is possible to achieve a reduction of the pit
geometry and thus, in principle, a higher storage density of
the disk by light of shorter wavelength. The first paper which
described the use of semiconductor lasers as write lasers was
published in 1981 by Jipson and Jones (IBM).147"1
They used
a hydroxysquarylium derivative (Fig. 10) as the recording
layer. which was applied to the substrate by sublimation.
This compound has good thermal and optical stability propAngel! Chem In[ Ed Engl. 2U (1989) 1445-1471
O n 0 OH
O H . . _1
O H ._.
Fig. 10. Syntheses for squaric acid and a hydroxysquarylium derivative.
erties. The pit formation proceeded in a reproducible manner
and is ascribed to pure sublimation processes.
However, such films are unsuitable for practical application since the sensitivity in the NIR range proved to be too
low and the long-term stability proved to be inadequate in
consequence of the tendency towards crystallization. In addition to this, the dye film did not demonstrate any threshold
value for pit formation, as a result of which, following frequent read out, a loss of the data was observed. Similar
results have been reported by Kivits et al.,'701 who used
recording layers consisting of vanadylphthalocyanine. The
number of NIR-sensitive dyes has greatly increased during
recent years. In practical terms, four categories of dyes have,
in particular, achieved significance: 1. methine dyes, 2.
phthalocyanine derivatives, 3. specific condensed arenes,
and 4.metal complexes. There are also many other examples
of dyes which do not fall into this subclassification; some of
these will be outlined after the discussion of the four classes
of substances.
4.3.1. Methine Derivatives
The first compounds of this category to be investigated
were heptamethine cyanines, with benzoxazole, benzothiazole, (benzoannelated) indole or quinoline moieties as terminal groups.[711In addition to having high absorption coefficients, these compounds exhibit excellent reflection properties (up to 40% at 800 nm) and thus yield materials with
good SNR values and high signal contrast. Indole derivatives are particularly preferred on account of their favorable
Cyanine dye films are almost exAngrw. Chwn. h i . Ed. EngI. 28 11989) 144s-1471
clusively produced by application from solution. In order to
coat the substrate with the dye, the latter must be soluble in
solvents which do not attack the substrate material. Propanol, ethyl acetate and-less preferred-acetonitrile have
proven to be suitable solvents. While the absorption maxima
of the monomeric methine dyes in solution are relatively
sharply localized, the half-width of the signals becomes
broader in a polymer matrix with the formation of new maxima. This broadening of the signal is ascribed to dye aggregat i ~ n . [ ~721
~ The
" . following properties distinguish the methine
dyes as absorbers for write-once disks: they have a high
molar extinction coefficient and require only small energies
to form sharp and intense pits (approximately 0.5 nJ/pit),
and a favorable signal-to-noise ratio results from the high
reflectivity ( > 30%). Methine dyes are photooxidized by singlet oxygen in the presence of light. The reaction is bimolecular and proceeds under unsensitized conditions relatively
slowly on account of small quantum yields; however, it is
accelerated by suitable sensitizers. The rate of decomposition
increases with increasing methine chain length, and may be
influenced by substituents in the heterocycle. The photodecomposition begins with 1,2-addition of ' 0 , to an exocyclic
CC double bond.[731
An increase in the stability may be achieved by the introduction of electron-withdrawing substituents, for example
halogens or cyano groups, as a result of which a slight
bathochromic shift of the absorption maximum takes place
in consequence of a reduction of the LUMO (Lowest Unoccupied Molecular
There are other possibilities
for stabilization by introduction of a cyclic unit (Fig. 1 1 ) into
the methine
It has been shown that particularly
H , Me
CH,, N R
Fig. 11. Stabilization of methine dyes sensitive to oxidation by introduction of
cyclic moieties. A = acceptor, D = donor.
suitable compounds are those derivatives which contain
~ q u a r y l i u r nor
[ ~croconium
like those shown in
Figure 11 (bottom right) in the methine chain. These have
been combined with the most widely varying terminal groups
such as a z ~ l e n i u m , ' ~p ~
y ]r y l i ~ m , [ benzo
~ ~ ] or naphthothiapyrylium,tsO1naphtholactam
or oxoindolizine heterocycles.[s21Furthermore, storage media consisting of polymeric
cyanine derivatives[831 or of tetra(dia1kylaminophenyl)cyaninests41have been proposed. The latter exhibit good solubility properties, but are inferior to the cyanines in their
optical properties. However, they are more stable towards
singlet oxygen, and the addition of radical traps increases
their stability.
The sensitivity to oxidation of theabove indicated methine
dyes may also be reduced by the addition of an oxygen
trap.[85J Examples of suitable compounds are dithiolato
complexes of platinum, cobalt and
(see also Section 4.3.4) or aminium derivatives,r871which can be obtained
as salts. They are added to the mixture or directly attached
to the dye.[881Alternatively, salts with a cyanine dye as a
cation and a radical trap as an anion can be used, e.g. a
dithiolato derivative. These salts tend to be highly ~ t a b l e . 1 ~ ~ 1
A particularly intensively investigated group of methine
dye derivatives comprises the squarylium dyes, which have
been developed by Philips and Kodak.r84d,84e1
The compound SQS (Fig. 12) which forms an inner salt can be spin-
\ /
ing stabilities, depending upon the environment. While water
and oxygen alone generated only very slight changes in the
optical properties of the recording layer, in the presence of
both water and oxygen peroxides were formed and relatively
large changes were observed. A reading test using a commercial NIR write/read laser revealed that after 10' reading
processes no change in the optical properties was detected
provided that a certain energy threshold value was not exceeded.184d.84%901
4.3.2. Phthalocyanine Derivatives
Phthalocyanine (H,Pc), a tetraaza[l8]annulene or tetrabenzotetraa~aporphyrin,~~~~
has proven to be a useful dye
for write-once optical storage disks. In Pc2@,four isoindole
units are cyclically connected to one another by nitrogen
bridges. The two protons of the metal-free substance may be
replaced by numerous metal atoms (MPc)C9'1 with valency
three or higher. In this case axial residues situated perpendicR'
A hrnl
Fig. 12. Transmission (T) and reflection (R) spectra of a 90 nm thick film of the
squarylium dye SQS suitable for a WORM medium. X = relative transmission
or reflection [84d].
= 600-800nm
= Cua,VOB, Al (OAlkyl) 2", Si ( O A l k y l ) ~ G e ( O A l k y l
= H,LF,OAlkyl. CH2O-Alkyl
coated as a dye solution (approximately 50 g/L n-propanol),
on account of the tert-butyl groups which increase the solubility. At 800 nm (Kr ion laser), an SQS film having a thickness of 90 nm reflects approximately 20%, and transmits
approximately 10YO (Fig. 12). Transmission and reflection
are functions of the layer thickness; transmission decreases
exponentially with layer thickness, while reflections are at a
maximum at approximately 90 nm layer thickness in the case
of SQS dye. The signal-to-noise ratio is a function of the
irradiated energy.
A decisive factor regarding the suitability as archiving material is the aging behavior of the dye layer. When stored for
relatively long periods at 90 "C,pure SQS films tend to crystallize, leading to a change in the optical properties (e.g.
doubling of transmission after 20 days at 90 "C). This can,
however, be controlled by the addition of small quantities of
a polymer. At room temperature, the pure film also exhibits
virtually no change. In the Z/AD test (temperature change
between 25 and 65 "C with relative atmospheric humidity of
93 %), the transmission of SQS varies by a factor of 2 after
40 days. SQS is rather stable towards light; this was tested at
40 "C and 65% atmospheric humidity using white light having an intensity of 1000 Wm-*; the SQS films showed vary1456
Fig. 13. Peripherally substituted phthalocyanine derivatives for WORM media.
ularly to the Pc plane can be introduced leading to a considerable improvement in the solubility of the virtually insoluble original substancerg3] (Fig. 13). In the presence of
peripheral s u b s t i t ~ e n t s 951
, ~ ~the
~ , solubility in aqueous solutions or organic solvents is increased. Unsubstituted
phthalocyanines (Pc) are easily synthesized in high yield and
are compounds of high chemical stability. More important,
phthalocyanines are attractive as optical recording media:
they have high absorption coefficients E > lo5, they are thermally stable, they exhibit virtually no toxicity, they show an
excellent stability towards atmospheric effects and they may
be applied to the substrate by sublimation or, with appropriate substitution, by spin-coating. Their absorption properties are determined by the rigid, internal 18x electron system
as a chromophore. They absorb especially well in the range
of HeNe laser radiation (633 nm). In 1981, Kivits et aI.[701
were the first to propose the use of phthalocyanines as absorbers in storage disks of the WORM type and to expand
the range of spectral sensitivity of the particularly suitable
vanadylphthalocyanine (VOPc), the absorption maximum
Angew. Chem. Inr. Ed. Engl. 18 (1989) 1445-1471
of which is at 730 nm, by thermal treatment;[961this results
in a broadening of the long-wavelength absorption band
beyond 800 nm. Using this material, it became possible to
produce a recording medium with a weak SNR with the use
of a krypton laser (799nm). A bathochromic shift and
broadening of the longest-wavelength absorption maximum
of phthalocyanines, e.g. in the case of MgPc, AICIPc,
I ~ C I P C [ ”or
’ ~T ~ C I P C , may
[ ~ ~ ]also be achieved by a THF or
CH,CI, solvent or vapor treatment.
It is possible to achieve a bathochromic shift of the
longest-wavelength absorption band by the use both of
strongly electron-donating and -withdrawing substituents.
Thus, in particular, hexadecafluoro derivatives of Cu2@,
ZnZ@,Pb2@or SnCI:@ show high absorptions above 800 nm
and are suitable for irradiation with semiconductor
lasers.[”’l Similar structural modifications, for example by
the introduction of peripheral azo or thiol radicals, have
been documented in numerous patent applications, predominantly by Japanese cornpanies.[loo1
Apart from these few exceptions, the influence of peripheral substituents on the absorption behavior proved to be
minimal. However, if the conjugated system is enlarged by
benzoannelation to give the naphthalocyanines (Nc)
(Fig. 14). then a substantial bathochromic shift of the long-
The naphthalocyanines are even less soluble than the
phthalocyanine compounds, and in most cases cannot be
sublimed. In order to increase their range of application,
soluble naphthalocyanine derivatives were therefore tested
for their suitability as absorbers for write-once disks; this is
possible, as with the MPc compounds, by peripheral or axial
substitution. The solubility is increased by adding peripheral
substituents, e.g. the bulky tert-butyl
there have been reports of numerous naphthalocyanine
derivatives which obtain their solubility by peripheral voluminous or long-chain alkyl radicals,[’ O4] carboxylic acid esters of long-chain alcohols[’051or long-chain polyethers.[’061
The biaxial derivatives of silicon-naphthalocyanine
(SiNc2@)(Fig. 15)11071have been noted for their economical
~ NJ“
c;H31CH2)16 CNH (CH2)fjO-Si - 0- - SiL0-Si
-0 ICH2)sNHC (CHz),rCH3
/ N
Fig. 15. Bisaxially substituted silicon-ndphthalocydninefor WORM media.
720 -820 nm
= Si.Ge
= HI t-C/+H9.(0CH, - CH~),,-OAlkyl,COOAlkyl
Si (A1kyl)3. Si (Alkyl I2-O -AI ky I
Fig. 14. Peripherally and axially substituted naphthalocyanine derivatives for
WORM media.
wavelength absorption band may result. While the absorptions of the 1,2-naphthalic acid derivatives show, in this case,
a slight bathochromic shift by 20nrn,[’O’’ in the case of the
2,3-naphthalic acid derivatives shown in Figure 14 this effect
is significantly more marked: displacements by more than
100 nm to 752 (PdNc) to 855 (MnNc(0COCHJ) have been
observed.[’o2] However, by enlarging the z-system, the
derivatives are less thermally stable and considerably difficult to synthesize.
A i q e n . C‘hetii. lnr. Ed. Engl. 2X (1989) 1445 1471
synthetic properties as well as their outstanding optical properties. The axial substituents can be selected so that they
guarantee adequate solubility as well as representing a type
of polymeric matrix preventing the tendency towards crystallization or reagglomeration. Compounds of this type exhibit high absorptions in the range of 780-830 nm, a reflectivity exceeding 30% and a signal-to-noise ratio of
> 50 dB.[107b1
The use of a soluble, axially substituted aluminum-naphthalocyanine has been reported; in this case, at
a write wavelength of 830 nm a sensitivity of 0.7 nJ/per pit,
a reflectivity of 25 %, a signal-to-noise ratio of 55 dB and a
reduction of the signal-to-noise ratio after 1.5 x lo6 reading
cycles by less than 1 dB (1 mW readout energy at 830 nm)
have been observed.[’081The derivatives must be extremely
pure in order to reproduce signals without considerable output loss. The most recent patent applications claim naphthalocyanine derivatives for WORM memories, which are
both axially and also peripherally substituted.[1091The synthesis of the naphthalocyanine system by peripheral substitution becomes very expensive since the corresponding 2,3dicyanonaphthalene derivatives must be produced as intermediates. The expenditure and the raw material costs may be
reduced by the use of phthalonaphthalocyanines.[l’ol
The naphthalocyanine derivatives represent an interesting
alternative to methine dyes as absorbers of diode laser radi1457
ation. Initial prototypes of WORM disks based on naphthalocyanines have already been put on the market.
4.3.3. Quinoid Polynuclear Aromatic Compounds
A further category of dyes to which reference is frequently
made in the literature and which can be used as an absorber
layer for WORM storage disks comprises the quinoid
polynuclear aromatic compounds. The use of these compounds is based on a different concept: the dye molecules are
rather small, generally of low solubility and, in contrast to
the methine dyes, uncharged, so that they may be applied to
the substrate by means of vapor deposition techniques
(Figs. 16 and 17).
Although this method of coating is more complex and, in
general, more costly, it also has its advantages. For example,
where sublimation is carried out, any dissolution of a polymeric substrate is avoided, and the possibilities of providing
the substrate with a polymeric intermediate layer to improve
the total properties increase.
The most widely investigated group of dyes are amino
derivatives of the 1,4-naphthoquinone type (Fig. 16), which
have already been known for a long time in the dyeing industry as pigments with a broad color spectrum.["'] In terms of
their reflectivity and sensitivity properties, they are less attractive than the known cyanine dye systems but offer higher
chemical stability and give storage media with long service
life of the data. The systematic examination of 1,4-naphthoquinone derivativesr1121absorbing in the NIR range was
extended by Matsuoka et al. to anthraquinone, phenothiazinequinone and phenoselenazinequinone derivatives (Fig.
17)" 1 3 ] and previously developed synthetic strategies by
means of PPP molecular orbital computations.[1141In this
case, it became evident that the long-wavelength absorption
maximum was influenced both by a change in the donor- and
in the acceptor-moieties, since the coloration of these compounds is generated by an intramolecular charge transfer.
Derivatives of the type shown in Figure 16 contain the com-
they have the disadvantage of high output losses on vacuum
An improvement in the chemical and thermal stability
(and thus in the sublimation properties) has been achieved by
the phenothiazine or selenazinequinone derivatives (Fig. 17)
and their isomers.[1161These compounds form homogeneous
X = S, Se
= 730 nm
= s, Se
X = H, Br
= 720 - 800 nrn
= 650 - 850 nm
Fig. 17. lnfrared absorbing phenothiazine and phenoselenazine derivatives
with benzoquinone (a), naphthoquinone (h) and 1.4-diketo units (c).
films with smooth surfaces.["'I Their absorption range is
between 750 and 850 nm, and strongly electron-withdrawing
substituents have a strong bathochromic effect." "1 Structurally related compounds used include specific indanthrene
and amino-substituted phthaloylacridones."20'
4.3.4. Metal Complexes
While the phthalocyanine ligand (Pcz0) is itself a chromophore, there is a series of colorless, in most cases sulfurcontaining ligands which absorb in the visible or near infrared range only as a result of 2: 1 complexing by means of a
metal ion. The long-wavelength absorption occurs due to the
formation of a 10 n electron system in the complex. A typical
example is represented by the dithiolatonickel complexes
(Fig. 18), which can be isolated as salts o r neutral complexes
NH 0
885-1165 nm
R = H, CH,,CI, NMe,
715-780 nm
R = H, CF,, Alkyl
Fig. 16. S,X-Diarnino-l,4-naphthoquinones
absorbing in the near infrared
bination of a suitable donor and strong acceptors. In solution, they absorb at approximately 770 nm, with high molecular extinction coefficients, and respond to the radiation of
commercial laser diodes. At a signal-to-noise ratio > 50 dB,
an acceptable reflectivity of up to 25% was observed at
830 nm.['151 The substitution of the free amino group by
alkyl o r aryl radicals leads to a further bathochromic shift.
In 1985, NEC announced a WORM storage disk based on
the ethyl ether (Fig. 16, R = C,H,). Although these ethers
exhibit optical properties which are on the whole acceptable,
= 850- 1300 nm
R', RZ = H, C1, Me, OMe, NMe,
Fig. 18. Infrared absorbing dithiolatonickel complexes
and exhibit absorption bands between 715 and 1300 nm in
solution.[' 2 1 1 Other suitable metals are palladium, platinum
o r alternatively cobalt, copper, zinc and cadmium. In DIP
Angew. Ckem. In[. Ed. Engl. 28 (1989) 1445-1471
configurations, an additional bathochromic shift by 20 60 nm is observed, which limits the use of numerous compounds of this type in WORM storage
The solubility of the easily accessible original substances
in organic solvents is moderate, but can be increased by
appropriate substituents so that they can be spin-coated
from solvents. However, the reflectivity values observed are
distinctly inferior to those of cyanine dyes. For this reason,
they probably will not be used alone as absorbers in WORM
storage disks.
It has already been pointed out, in connection with the
cyanines, that dithiolato complexes are very efficient oxygen
traps and thus considerably increase the stability of cyanine
absorber layers. In 1985, almost 60 patent applications were
filed in which such combinations are claimed. At the present
time, there is no precise concept concerning the protection
mechanism. Closely related classes of compounds include
s1, a-diimino-cis-I ,2-ethylenedithiolatonickelderivatives
or complexes in which aromatic o-diamines, o-aminothiols
or o-selenols (i,
quinolinediones"271 are used as ligands.
4.3.5. Other Dyes
In addition to the most important classes of compounds
which have been outlined in Sections 4.3.1 -4.3.4, the patent
literature also discloses a whole series of other NIR-absorbing compounds. These include specific aminoarylpentafulvenedicarbonitriles (i.,
680-785 nm),[1281ethynevinyl
di- and triphenylcyclopropanes (2,,, = 530- 820
and triquinocyclopropanes (
770 nm)," 301 which can
be applied by sublimation or, with appropriate substitution,
by spin-coating, polypyrrole (1 nJ/pit at 830 nm)I13'l and
intermolecular charge transfer complexes with 5-nitro-2,3dicyanonaphthoquinone as acceptor (lmaX
< 740 nm).11321
5. Reversible Data Storage
The set of requirements imposed on erasable optical data
stores is prescribed by the current specifications of the conventional magnet technology (Winchester disk drives), since
the two technologies are designed for the same range of
application. The storage densities of the EDRAW media
which are in an initial marketing phase, currently amount to
approximately 10' bitcm-', while hard disks on average
achieve lo6 bit cm-2.1133-1351
An increase in the bit density
of magnetic hard disks to lo7 bit c m - 2 is under investigation
by the use of special preparations of y-Fe,O, o r new types of
cobalt alloys.[141,1491 The direction of magnetization for a
recording is not altered in the plane of the disk (longitudinal), but perpendicular to it so that magnetic domains
< 1 pm can be in~cribed!'~~1
In magnetic high-performance disk drives, the aerodynamic write/read head is guided a t a height of approximately
0.1 p,l'"l
and therefore it is possible for "head crash" to
occur as a result of dust particles and vibrations. The diskdrive must therefore be mounted rigidly as well as protected
from dust and other foreign particles. In contrast in the
optical process, the corresponding components are situated
a t a non-critical spacing, which gives substantial insensitivity
Clieiii. h i .
Ed. Engl. 28 (1989) 1445- 1471
to particles, as well as no encapsulation of the disk and. in
consequence thereof, an exchange facility and secure handling.
In optical data carriers, the density of the guide tracks
(8000-9000 tracks cm- I ) is higher than with magnetic carriers (1000-1 500 tracks cm-1),[138]so that the storage capacity of optical systems is greater by a factor of 10 than that of
the Winchester disks."39. 1401 On the other hand, data access
via hard disks is considerably faster (8-20 ms) than with
optical disks (30-50 ms) and the data rate, approximately
12 Mbyte s-', is two to three times greater than with optical
recording processes. By increasing the speed of rotation to
4000 rpm and using a laser diode array, the data rates for
optical media may be increased to 6 M b y t e s - ' or
> 10 Mbytes-'.[142,1431It is expected that by about 1995
the access times may even be reduced to a value of 20 to
25 ms.11441General requirements to be imposed on reversible
storage media include bit error rates of approximately 10(after correction), application cycles exceeding I O7 and the
"direct overwrite" of stored information. In a similar way to
the write-once memories, a defined threshold value or a steep
progression of the energy/sensitivity curve is required for the
reversible types of marking of the EDRAW media. Other
conditions include long-term stability of up to 10 years for
the archiving of information and compatibility with existing
hardware and software.
Several different materials may be used for reversible data
storage, but only a few are at the stage of industrial application. It is required that the information can be easily read
and erased, and this leads to a conflict which can be resolved
only in circumstances in which a material at room temperature exhibits two conditions which can be selectively addressed and switched. However, many materials which are in
principle appropriate are altered in their condition of marking when a laser beam is frequently used for readout.
The development of reversibly operating magneto-optical
memories and phase transformation media has progressed to
a stage a t which marketing is imminent. The overview given
below therefore concentrates on a description of these processes.
5.1. The Magneto-Optical Effect
The magneto-optical (M/O) effect can be used for the
readout of data, since a rotation of the plane of polarization
takes place as a result of the interaction of a polarized light
beam with magnetic material (Fig. 19).[1451In this way, it is
possible to read data markings in reflection (M/O Kerr effect) or in transmission (M/O Faraday effect). Writing and
erasure by means of a laser beam is based on the temperature
dependency of the magnetization (Fig. 19). A recording
medium magnetized perpendicular to the disk is heated at
definite positions, and the magnetization is flipped over into
the opposite direction by a magnetic field. There are three
categories of recording materials used : crystalline intermetallic compounds, amorphous rare earth/transition metal
alloys (RETM), and ferrites.
The progression of the magnetization of alloys of rare
earths and transition metals is shown as a function of the
temperature in Figure 20.1'01The RETM alloys are included
M-0 medium
angle of polarization
of laser beam
altered by Kerr
Fig. 19. Principle ofmagneto-optic (MO) data storage [145]: a) the perpendicular magnetization of the MO-medium is unaffected by magnetic field a t room
temperature; b) recording via heating by laser beam and turnover of the magnetization via a magnetic field; c) reading via Kerr effect; d) erasing via heating
by laser beam and turnover of the magnetization according to original direction.
among the ferrimagnetic substances. The magnetic structure
is characterized by the presence of various subunits, which
are magnetically coupled and assume an antiparallel orientation. The temperature dependence of the magnetization of
the subunits differs, so that the difference represents the
spontaneous magnetization of the entire system. In this case,
the orientation is perpendicular to the plane of the recording
medium (perpendicular magnetic anisotropy). At low temperatures, the magnetic moment of the rare earth atoms is
greater; at a higher temperature, the magnetic moment of the
transition metal atoms predominates. Accordingly, as a result of the antiparallel configuration, the temperature progression of the total magnetization is characterized by two
temperatures. The point of compensation is marked by the
temperature (~,,,,)
a t which the magnetizations are equal,
and the net magnetization of the alloy becomes zero. In the
event of an increase in temperature, the magnetic moments
of both species of atoms disappear at the Curie point (T,,,,,)
and the alloy material becomes paramagnetic from this temperature upwards.
Fig. 20. Temperature dependence of magnetization M and coercitivity H, of an
alloy RETM of rare earth element (RE) and transition metal (TM) [lo].
The preferred direction of the total magnetization is given
by external magnetic fields. The stronger the external field
and the lower the temperature, the more completely the external magnetic field forces the orientation of the elemental
magnets in the field direction. O n account of the high perpendicular anisotropy, the maximum attainable magnetization, i.e. the saturation magnetization, can be changed only
in circumstances in which the external magnetic field reaches
a threshold value, i.e. the coercive field strength ( H J . At the
compensation temperature, the coercivity of a ferrimagnetic
alloy is theoretically infinitely large, so that due to large
magnetic coupling, the direction of the magnetization cannot be altered by external fields. ~ b o v e~c~~~~ the coercive
field strength falls again to small values and at the Curie
point becomes zero.
Depending upon the writing temperature, a distinction is
made between compensation point or Curie point writing.
For compensation point writing, Tcomp
is set by the composition of the alloy to room temperature. The writing process
takes place with simultaneous heating of the recording medium by a laser beam and application of an antiparallel external field by means of the disk drive magnet. As a result of
heating, the coercive field strength diminishes and the direction of magnetization is flipped over in the heated area-the
data bit-by application of an external field and read out by
means of the Kerr effect. Curie point writing involves the
heating of the metallic layer to achieve the elimination of the
magnetic order at Tcurie
and setting of the new preferred
direction by an external magnetic field. The erasure process
is distinguished from writing only by the direction of the field
of the disk drive magnet.
In comparison with the magnet technology, the smaller
data rates of the M/O technology are determined by an additional disk rotation, which is required for erasure, in order to
create the orginal magnetic orientation on the disk. Thus, a
total of two disk rotations must be executed for the writing
of new data. Accordingly, M/O disk drives of the next
generations will be provided with a direct overwrite
deviCe,[146.147, 1491 0ne possibility is the use of a magnetic
head, the field of which switches the data spots heated by the
laser in accordance with the bit stream (Magnetic Field
Modulation Method). In this way, data rates exceeding
10 Mbyte s- have been
The distance between
the head and the disk is 2-5 pm, so that conditions are not
as critical as with the Winchester disk drives.
The layered structure of a storage medium is of prime
importance for the optical properties. However, the RETM
media adopt a certain special position, since they must be
encapsulated with protective layers on account of their sensitivity to oxidation. The protective layers fulfill several functions: a barrier against oxygen, water vapor, substrate, and
protective polymer lacquer, an antireflection layer for coupling in the laser light, and thermal insulation of the recording layer in order to keep the temperature of the substrate
constant. In most cases, use is made of multilayer structures
which, depending upon the construction and layer thicknesses, influence the maximum attainable signal intensity and
sensitivity. Thus, reports have been given, for example, not
only of simple systems consisting of an M/O layer, antireflection layer and metal mirror, but also of constructions consisting of a barrier film, two different M/O layers and an
antireflection layer.@*231 In this way, the reflectivity of a
RETM medium may be varied between approximately 4 and
30 %. The materials used for barrier layers o r dielectric layers are in most cases amorphous films of SiN, AISiN, or
BN." 531 Additionally, investigations have been carried out
on TiO, TiON, TiN, A N O N , S O , AIN, SiON, ZnS and
SiO,.I1stl For the setting of an optimal SNR, the materials
Angcw. Chrm. Int. Ed. EngI. 28 (1989) 1445-1471
of barrier layers must exhibit amorphous structure, low birefringence, high transparency at the laser wavelength, low
thermal conductivity, and must by easy to manufacture. The
Kerr angle can be increased by appropriate selection of refractive index and layer thickness of the barrier film."
A prerequisite for the magneto-optical data storage is a
perpendicular magnetic anisotropy of the recording medium. The causes of the anisotropy have, to date, not been fully
clarified; it is generally assumed that three phenomena may
occur: spin-orbit coupling, magnetic dipole interaction and
1 7 7 3 2 2 2 - 2 2 5 1 However, it has been verified that a relationship exists between magnetic anisotropy
and the process of preparation of the RETM layer. Amorphous RETM alloys can be produced by a multiplicity of
vacuum coatings:
and electron-beam vapor dep ~ s i t i o n , ~ 'ion-beam
s p ~ t t e r i n g , ~ ' ~R~F~ ' , sputtering,""']
and DC magnetron
have all been
used with success. Certain properties of the materials can be
influenced by the coating process to a far greater extent than
by controlling the layer
M/O films consisting of RETM alloys are produced by DC magnetron sputtering for the production of data stores.['571
drops. A difference of 2 atom-% in the cobalt content causes
by approximately 50K['88J(Fig. 21). It is
a change of Tcurie
possible to produce homogeneous GdCo films by applying a
negative bias voltage to the substrate.[I6" 1 7 4 1 It is assumed
that the oxygen content of the film is reduced by increased
argon ion bombardment.['691 In this case, an anisotropic
distribution of Gd and Co is generated by selective resputtering of gadolinium atoms, and thereby the perpendicular
magnetic anisotropy is built up as a function of the bias
~ o l t a g e . ~A' ~columnar
structure is generated in the layer
by the processes taking place during sputtering." 7 6 1 In other
investigations, gadolinium was replaced by holmium['641
= 46°C) with the composition Ho,,Co,,. The coercive field strength of HoCo films is greater than the comparable GdCo layers, and it was possible to write stable data
5.2. Magneto-Optical Materials
The interest in M/O materials for data recording dates
back to work by Williams in 1956 in the Bell Laboratories;
a hot needle was used to inscribe magnetic domains into a
manganese/bismuth film which were read out by means of
the Faraday effect.['s81 Curie point writing using an electron
beam was demonstrated in 1958 by Mayer,[1s91
and the Kerr
effect was used for the first time for the detection of magnetic
domains in 1960.['881Compensation point writing was attempted in 1965 on Gd,Fe,O,,, a magnetic garnet.['601
Lasers were developed parallel to this work and gained increasing acceptance as a radiation source for writing, reading
and erasure. These first M/O materials exhibit a series of
properties which prevented large-scale technical use as storage media['6'.1621and led to the development of the RETM
alloys, which are today the most important materials for
M/O memories.
Chaudhavi et al. were the first to show that gadoliniumcobalt and gadolinium-iron films exhibit spontaneous perpendicular magnetic anisotropy." 6 3 1 The variation of the
compensation point between -233 "C and +230°C was
achieved by the selection of the Gd, -,Cox composition. At
the compensation point the coercivity showed a maximum,
which was greatly dependent upon the method of production. In initial writing attempts at the compensation point, it
was possible to inscribe data markings. In further investigations, it emerged that the stability of small magnetic domains
is very low.
A fundamental disadvantage of GdCo resides in the difficulty of obtaining homogeneous layers with a constant compensation point. The compensation point decreasing from
the center towards the edge is based on irregularities of oxygen traces in the layer.['69-'731 In terms of magnetic properties, oxidizied gadolinium behaves differently from pure
gadolinium ; accordingly, the relative proportion of cobalt is
increased, and as a result of this the compensation point
A n x w . Chrin.h1.Ed. Engl. 28 (f989) 1445 -1471
RE content TM [atom %I
Fig. 21. Curie temperature T, as a function of rare earth content of GdFe.
TbFe, and DyFe alloys and compensation temperature
of GdCo as a
function of Gd content [188].
In Gd alloys, the replacement ofcobalt by iron leads to the
lowering of the Curie temperature (T,,urie)['7*1801 and Kerr
angle."*'] The coercivity of GdFe alloys is, like that of
GdCo, relatively
The properties of GdFe compounds can also be modified by alloying with other metals:
yttrium increases H , and reduces Tcurie;
and lead['871increase the Faraday and Kerr effects; bismuth
and tin cause the lowering of Tcurieand
l S 6 l If
gadolinium is used in place of terbium in REFe alloys, materials of which the Curie temperatures are approximately
100K below those ofGdFe['**. 189i(Fig.21) are formed. On
going from gadolinium to terbium to dysprosium, the progressions of the TcUri,
curves are displaced in almost constant
intervals by 50-100 K in each case. The Curie temperatures
of iron alloys are significantly less sensitive with respect to
variation of the rare earth element content (Fig. 21).
In the case of the iron and cobalt alloys of gadolinium,
compensation temperature writing is employed; in the case
of terbium and dysprosium alloys, Curie point marking is
mostly used. On comparing coercive force and spontaneous
i F (
Fig. 22. Principle of MO layers for recording of small domains: a) double layer of TbCo alloy with C o enrichment in the upper and T b enrichment
in the lower film. The arrows indicate the preferred directions of the magnetic moments of the T b (+) and Co atoms (a)
as well as of the resulting
T'o,,,p(I)< room tempermagnetic netto moment of the layer (+): b) two GdCo layers (I and 11) with compensation temperatures Tco,,,~(Il)
ature; c) small magnetic domains in GdFe via exchange coupling with a nickel layer.
magnetization as a function of the temperature, similar
marked differences are found as for the Curie points in the
case of the iron alloys with the heavier rare earth metals
(RE = Gd, T b and Dy). GdFe alloys exhibit the already
mentioned low coercive field strengths, which lead to instabilities on writing small bit domains. TbFe and DyFe are
characterized by coercive field strengths of a few
100000 A m - and steep temperature dependence of the
magnetization; this is desired for the writing of small domains.
The methods by which the TbFe films are prepared determines to a great extent their properties. Uniaxial anisotropy
K, and saturation magnetization of TbFe films can be varied
as a function of the substrate
in contrast,
the perpendicular anisotropy of GdFe is almost independent
of the substrate temperature. Using X-ray scattering, it was
found that during the coating an oxidation of terbium to the
suboxide takes place, which causes, at a relatively high temperature, the large drop in the saturation magnetization in
terbium-rich films.
The magnetic and magneto-optical characteristic quantities of TbFe layers can, moreover, also be modified by the
variation of the film thickness['911 and the pressure of the
sputtering gas,['921and can be optimized by selection of a
suitable layer construction with respect to their dynamic
write/read/erase properties.['931
The mechanical conditions applicable to the substrate material can also have an effect on the magnetic anisotropy
constant, and can generate orientation effects by anisotropic
stresses (magnetostriction), for example in foils as carrier
material for magneto-optical layers. Attempts at coating
TbFe films onto Kapton foils by R F sputtering led to recording media with a high magnetic anisotropy, which could
possibly be used as M/O tapes.['941
The RETM alloys are particularly oxygen-sensitive and
must be protected from the atmosphere for long-term storage. Within the individual alloy systems there are, however,
significant differences. RECo films are far less subject to
oxidation than REFe
and exhibit higher thermal
~ t a b i l i t i e s . ~ The
' ~ ' ~write sensitivity is very high in the region
of the compensation composition on account of the sharp
decline of H c , so that it is possible to write using lasers of low
energy; therefore the stability is greater upon frequent read-
Alloys of terbium and cobalt have been tested; unlike
GdCo, these alloys are more stable and d o not contain inhomogeneities of the layer when gadolinium is replaced by
terbium. Using homogeneous TbCo films, it was not possible to achieve
by modulating the terbium/cobalt
ratio within a film, it was possible to increase magnetic coercivity and resistance to oxidation.['961 Using bias R F sputtering, cobalt-rich films have been produced at a bias voltage
of 100 V and terbium-rich ones without a bias (Fig. 22a). At
a negative bias voltage, it is assumed that terbium is preferentially resputtered, so that a relative increase in the cobalt
content results. The lattice positions which have become free
in consequence of terbium resputtering are clearly occupied
by the sputtering gas; this leads to a more porous microstructure and easier oxidizability than in the case of films sputtered without. In this way, films with up to 36 layers may be
prepared; these have an average thickness between 80 and
100 A. Between the layers, compensation zones are formed
which result in a homogeneous layer only from a film thickness of approximately 40 A. By appropriate selection of the
layer composition of the terbium-rich layer, the compensation temperature has been set above room temperature so
that in consequence of exchange coupling of the magnetic
moments between the two layers, a high coercive field
strength (ca. 320000 A m - ' and above) is achieved. The
Faraday rotation of TbCo media read out in reflection is
mainly based on the cobalt sublattice and is less due to the
influence of terbium,['971in contrast to the TbFe alloys, in
which both components contribute to the Faraday eff e ~ t . ~ ' ~Effects
* I of the exchange coupling of heterogeneously
structured RETM films have also been investigated on
GdFe/DyFe and GdFe/TbFe layers.['841
In many cases the requirement for high write sensitivity is
incompatible with the requirement for high domain stability,
bit density and a high S N R both for TCuri,
and KO,, writing.
By decoupling the writing and reading processes onto two
separate layers these weak points are avoided.['841The write
film should ideally exhibit high coercive field strength at a
low Curie temperature in order to achieve high sensitivity
and domain stability. The read layer must be coupled by
means of exchange coupling to the spin moments of the write
layer and also exhibit a high Curie temperature and coercive
field strength for the transfer of written domains from the
write onto the read layer. TbFe and DyFe layers were tested
as write films in conjunction with deposition of a GdFe layer
as read film.['841In the case of TbFe/GdFe structures, coercive field strengths increased to a few 100 000 A m-I, and an
associated domain stability of up to approximately 1 pm
were found. Accordingly, the properties of monofilms consisting of GdCo or GdFe were exceeded.
Chein. hi.Ed. Engl. 2X (1989) 14451471
Another method for writing small magnetic domains consists in the use of two layers of G d C o with differing compositionI'65-1h71(Fig. 22 b). The compensation temperature
7&, of the upper layer (I) is set by the ratio of G d to Co to
be somewhat below room temperature, and Torn,
of the lower layer (11) is below T,,,, of the upper film. This means that
at room temperature the saturation magnetization M , and
the demagnetizing field Hd of the upper layer are small (the
field lines of an upwardly directed magnetic domain penetrate the surrounding layer in the reverse direction and therefore act in the sense of a demagnetizing field), while M , and
Hd are greater in the lower layer. In this way, domains of the
upper film are stabilized. This also applies to the case in
which the preferred direction of the anisotropic magnetization is not exactly vertical, but tilted.['681
Small magnetic domains in the pm range can also be written with the aid of a layer of ferromagnetic material['83, l S 4 ]
(Fig. 22 c). It is assumed that by vapor deposition of a nickel
film on the G d F e recording layer a written magnetic domain
is stabilized by the formation of a locally closed domain in
the longitudinally magnetized Ni film. The coercivity of the
entire GdFe-Ni structure becomes greater and a higher magnetic field strength is required to switch the data bit.
Ternary and quaternary alloys of rare earths and transition metals open up a multiplicity of design possibilities for
structure/property relations. The search for technically suitable M/O materials extends on a preferential basis to alloys
of such transition elements as iron and cobalt with the heavy
lanthanoids gadolinium, terbium, dysprosium, holmium, erbium, neodymium, praseodymium and samarium" OS 1891
(Fig. 23a). In addition, films of the most widely varying
(Tb-Felloo-y X,
= Sm
A Dy
0 Ce
y [atom-%]
ternary and quaternary alloys have been described: with
cerium, promethium and europium as rare earth elements," 891 and titanium, vanadium, chromium, manganese,
nickel and copper as transition metals[1991(Fig. 23 b), as well
as various doping elements such as bismuth,['851 tin,[zoo1
molybdenum, gold,[z021 silver,
palladium, platinum [ I g 9 ] and uranium.[203"]In Figure 23 the
Kerr angle 9, of some ternary Tb-Fe alloys are shown as a
function of the content of the third alloying component.
The number of the most promising alloys is, however, for
marketable products, relatively small, and restricted to the
GdTbFe, TbFeCo systems and their four component alloys
with dysprosium and neodymium respectively. By using
ternary GdTbFe alloys, it is possible to combine the properties of TbFe and GdFe:[2061the Kerr angle (ca. 0.30') is
about the same (TbFe 0.25", G d F e 0.30"; at 1. = 800
nm),[1551the Curie temperature (between 150- 180 "C) is
reduced as compared to G d F e (200-250 0C)J2041and the
coercivity is increased.[203b1
Dynamic memories and reading experiments with
GdTbFe films[205-2071demonstrate the usefulness of this
ternary alloy system. In initial experiments, bit error probabilities of 10-3-10-4 and lo7 intact write cycles were obtained.[2071The magneto-optical rotation was optimized using different metal reflection layers (Au, Cu, Ag) and
By using a quaternary alloy of
dielectric films (Si02).12081
GdTbFe with dysprosium, it was possible to increase the
signal from 0.7 (at 800 nm) to 1.7 (at 700 nm). By replacing
gadolinium by dysprosium in ternary RETbFe alloys, the
Kerr angle (Fig. 23a) and the Curie temperature are reduced.[2081The Kerr effect can be increased by a three-layer
structure of Cu reflection layer, DyTbFe, and SiO, dielectric. The write process was executed with an 8 mW semiconductor laser. The optical transmission of the write/read/erase
head showed approximately 4 4 % power loss, so that it was
possible to measure 4.5 m W write power and 2 rnW read
power on the disk. The prerequisites for the writing of magnetic domains < I pm were satisfied by a coercive field
strength of ca. 80 000 A ni- ', and the erasure of data bits
could be performed by means of a 12 000 A m - ' magnetic
field.[2081The bit density of this system was quoted as being
1.5 x lo7 bitcm-'.
As shown for the binary REFe alloys, together with an
increase in the order number of the lanthanoids it was possible to detect a lowering of the Curie temperature, even for
ternary alloy systems (Fig. 24).Il6] This applies in the same
\ k\
(Tb-Fell,o_y X,
\ x:
m Ti
y [atom-%]
Fig. 23. Kerr rotation angle 9, of (FeTb),,, ,X, alloys as a function of the
composition. a ) X = rare earth element. b) X = transition metal [189].
Angcii. C'lwm. Inr. Ed. Engl. 28 (19891 (445 1471
nFig. 24. Variation of Curie temperature T,,,,, with substitution of rare earths
for Tb in (Tb, xYx)o.zFeo.8(curve A) and in (Tb, ,Y.),,(Fe,,Co, 1 ) 0 8
(curve B) where x is determined by the average number of 4f electrons in the
alloy; 0 < x < 0.1 [209].
way to iron and cobalt alloys; the Curie temperatures of the
cobalt compounds are generally higher. The particular influence of the cobalt is also shown in comparison with the other
transition elements. The curves in Figure 25 indicate the
change in the Curie temperature of the amorphous Tb-Fe-
[ Klatom-%]
Fig. 25. Variation of Curie temperature Tc,,,. with substitution of small
amounts of various transition metals (TM) in amorphous Tb, ,(Fe, .TM,),
(curve A) and in crystalline Fe, .TM, (curve 9). respectively; 0 < x < 0.1 [16].
transition metal alloys (A) and crystalline Fe-transition metal compounds (B) as a function of the transition metal (up to
10 atom-%) for several transition metals. Thus, with a fixed
lanthanoid content, the Curie temperature can be set by the
ratio of two transition elements.
The temperature dependencies of spatial arrangement,
electron interaction, spin moment and exchange coupling
give the interelations between magnetization, Curie temperature and Kerr effect.1209.2101
Figure 26 shows a plot of the
wide limits. Tb,,Fe,,Co,,
(RE-rich) and Tb,,Fe,,Co,,
(TM-rich) films were investigated for their use as storage
media and were characterized with respect to their magnetooptical parameter^.'^ 31 Both layers exhibit a Curie temperature of 200 "C; however, there are differences in the saturation magnetization and coercive field strength. The storage
layer with the larger iron content shows a higher saturation
magnetization than the terbium-rich film. The marked magnetic domains become larger with increasing laser power and
external field strength as a result of the dissipation of thermal
energy and the consequent profile of the coercive field
strength H, in the layer. The domain size can be computed
from the temperature progression in the film, assuming a
gaussian distribution for the intensity of the laser beam, and
to a first approximation is given by Equation(a) where
T ( r ) ["C]= 120 exp ( 2 r 2 / 0 . 6 Z+) 80
r [pm] is the distance from the focal point to the laser source.
This term is based on the assumption that at the focus the
Curie temperature of 200 "C is reached and the laser beam
has a radius of 0.6 pm.
With the aid of this equation and of the temperature dependence of H,, the distribution of the coercive field strength
H , ( r ) in the layer can be computed (Fig. 27). The domain
Fig. 26. Kerr rotation angle 9, versus Curie temperature T,,,,. for various rare
earth substituted FeCo-alloy films. I2091
Kerr angles of several rare earth element-substituted FeCo
alloys versus their Curie points. It is evident that for a magnetization which is higher at room temperature a larger Kerr
angle results. In the case of alloying cobalt with TbFe and
GdFe, the Kerr angles increase. The effect is particularly
marked in the case of TbFe: in a Tb,,Fe,, alloy, the Kerr
angle increases from 0.18" to 0.37" for Tb,,(Fe, , C O ~ , ~ )at, ~ ;
the same time, the Curie temperature increases 4 K/atom-%
co.I 2 111
TbFeCo films are used in the majority of commercial storage disksJ2' accordingly, they have been extensively investigated. The properties of a M/O layer consisting of TbFeCo
may be varied by the ratios of the components within very
domain size
-- --r--8
M/O-medium/; 0
4 x t
domain s i z e
Fig. 27. Domain formation model for MO recording. [213] The profil of temperature Tand coercivity H, are represented for a Tb-rich (Tb,,Fe,,Co,,) and
an Fe-rich (Tb,,Fe,,Co,,) film. The domain size in a Tb-rich film is smaller
than in an Fe-rich film due to differences in Saturation magnetization. The
vertical dashed arrows in the upper part of the figure demonstrate the width of
the laser beam. The horizontal dashed lines indicate the initial niveaus of Tand
H , of the films.
radius r, is then obtained from the balance of the magnetic
fields in the layer [Equations(b) and (c)] as radius of the
circle in which (d) applies. x is the magnetic susceptibility, po
the magnetic induction constant and M , the saturation magnetization. He,, is the external magnetic field which is applied
to switch the heated data spot and together with the magnetic field Hd generated by the recording layer must be greater
A n g w . Chem. Inl. Ed. EngI. 28 (1989) 1445-1471
than the coercive field strength H,(Y,), in order to mark a pit
(see Fig. 27).[2'31
In consequence of the high saturation magnetization of
the iron-rich layer, the demagnetizing magnetic field Hd of
the Tb,,Fe,,Co,,
film is greater than that of the terbiumrich Tb,,Fe,,Co,,
film. In the case of an equal external
magnetic field He,,, the progression of H c ( r s )in the iron-rich
layer is substantially broader due to the differing H , fields,
so that larger domains are written. This can be an advantage
when the storage disk is rotated at high speeds and the temperature profile generated in the recording layer becomes
narrower (high speed recording).
Media with a write characteristic comparable with that of
the terbium-rich film are particularly suited for high recording densities. The demagnetizing field Hd generated by the
magnetization of the magneto-optical film can, if the composition is suitable, become so large that the coercive field
strength of the heated data spot is exceeded and writing can
take place without an external magnetic field H,,,; this is
true, for example, of Tb,,Fe,,Co, and Tb,,Fe,,Co,,.
form of the data spots is dependent upon the ratio of the
individual magnetic fields on writing and erasure respectively and has an effect on the signal-to-noise r a t i ~ ; [ ~ ' ~ - ' ' ~ ]
heat dissipation in the layer structure of the storage medium
represents a significant boundary condition. The laser beam
propagates heat, and therefore heat sinks in the form of
insulating dielectric layers are introduced, which are located
between the metallic reflection layer and the storage mediThe dimensioning of such multi-layer structures
consisting of an antireflection layer, a storage layer and a
reflection layer as well as dielectric, requires considerable
optimization. The radial temperature distribution in magneto-optical media has accordingly been the subject of numerous investigations, especially from the point of view of influencing the SNR.[2'6J
The modulation noise Nmod
is the noise difference before
and after writing, which can be measured on scanning the
data markings and is a measure of the regularity of the magnetic domains. The smaller Nmodthe greater is the optimally
attainable SNR. As a result of an inhomogeneous temperature distribution (caused in consequence of fluctuation of the
layer composition or within the laser beam) in a heated data
spot, differences in the magnetization may arise between the
center and the edge zone. Under particular conditions, these
differences may become so large that instead of punctiform
domains annular ones are written. These irregularly shaped
data spots increase the modulation noise and reduce the
By way of analogy with the binary alloys, the SNR can be
increased by specific layer structures. TbFeCo was used to
produce layered films by means of sputtering, with which
high SNRs were
The optimization of the
signal-to-noise ratio, which is proportional to 9,
(R = reflectivity, 9, = Kerr angle), was described for threelayer storage systems using TbFeCo and DyFeCo alloys.[2
The most suitable layer thicknesses for the dielectric film are
those given at the largest possible Kerr angle and low reflectivity. Using (Nd,3Dy,,)30(Fe84Co,~)70
films, data rates of
10 MBytes-' for 5.25" disks at 50 dB for the SNR value
were achieved in dynamic write tests.[22 Moreover, investigations were carried out on quaternary RETM alloys of the
Angew. Cliczm. In!. Ed. Engl 28 (1989) 1442-1471
and GdXFeCo type,[2201where X = Dy, Ho,
and Er, in order to modify Curie temperature. coercive field
strength, Kerr angle and magnetic anisotropy.
In contrast to the amorphous RETM alloys, ferrites are
air-stable M/O materials which are capable of being used
without costly protective layers. Recently, these oxides
gained renewed importance after having lost significance for
reasons associated with the difficult process technology and
inadequate quality of reproduction.[2391Ferrites are mixed
oxides consisting of trivalent iron and various metal cations
which crystallize with spinel, garnet or hexagonal type structures.
In the field of data stores, hexagonal ferrites have been
recognized as hard magnetic pigments for the magnetic technology.'2261As far as M/O storage is concerned, barium
ferrites BaFe,,O,, with a magnetoplumbite structure are of
special interest. The barium ions can be replaced by strontium, lead or in some cases calcium ions, or alternatively by
equimolar quantities of M' (Na, K, R b etc.) and M"' (La, Pr,
Gd etc.). There are five magnetic sublattices with differing
spin orientation of the Fe3@ ions. At a wavelength of
788 nm, BaFe,,O,, shows a Kerr rotation of 0.22" which
increases on partial replacement of the Fee by Co2@/
The coating of the substrate with a ferrite film can
take place by epitaxy from the meltt2281
or by reactive sputtering on glass disks at substrate temperatures above
550 OC.[Z291
Ferrites having a spinel structure have the general composition MFe,O, (M = Co2@,Mn2@,Ni2@,Zn2@etc.). At
700 nm, chromium-substituted and rhodium-substituted
cobalt spinels CoM;Fe,-.O,(M' = Cr, Rh) exhibit Kerr angles of approximately 0.3 -0.5". The Curie temperature of
the cobalt spinel (x = 0) is 520 "C. The very high rate of heat
loss, which arises in the course of the Curie point writing of
thin layers of these materials, can be reduced by incorporation of a dielectric layer into a three-layer structure of the
storage r n e d i ~ r n . I As
~ ~ ~a ]result of the incorporation of
chromium or rhodium, the Curie temperature of the cobalt
spinel is reduced: Mi = R h , , T, = 62°C; Mi = Rh,,,,
= -113°C; M: = C r , , T,,,,, = 107"C.[2301On account of the high saturation magnetization, domains are
marked at Tcurie
without an external magnetic field. In order
to improve the magneto-optical properties, the combination
of ferrites in mutually superimposed film layers was also
tested, and, for example, using NiFe,O, and C O F ~ , O , , [ ~ ~ ~ '
it is possible to produce aluminum-cobalt ferrites
CoA1,Fe,-xO, as film by sintering at 1250°C; as the aluminum content rises, the Curie temperature falls from approximately 500°C (x = 0.1) to 150°C (x = 1.0).[2331
Ferrites having a garnet structure--M,Fe,O,,
(M = Y,
Gd, Dy, Ga, Tb; M, = M'M"M"' )-are similar to the rare
earth/transition metal alloys. 'In garnets of the type
Gd, -,Bi,Fe,O,, , a displacement of the compensation temperature was measured as a function of x; A Komp= 135 K
per atom-% Bi.t2341The sublattice of the Gd3@ions couples
antiparallel to the Fe3@sublattice. At low temperatures, the
moment of the rare earths is greatest, while at higher temperatures that of the iron ions is predominant. In the middle
temperature range, a compensation temperature occurs.
In a similar way to the case of the RETM alloys, the
magnetic anisotropy constant K , is dependent on the condi1465
tions of the film preparation, which can vary widely when the
layers are prepared by liquid-phase epitaxy on the one hand
and by sputtering on the other.[2351In spite of their outstanding physical and chemical stability, magneto-optical films
consisting of ferrites are at present not yet suitable for industrial application in commercial products since the Curie temperatures are too high and light absorption and surface
smoothness are too low. As a result of the crystalline structure, the signal noise is very high compared to that of the
amorphous RETM materials. The possible coating methods
for magneto-optical oxides require high temperatures
( > 500"C), so that the use of polymeric substrates demands
the development of special processes. (GdBi),(FeAIGa),O,,
could be sputtered by using a garnet target while maintaining
a substrate temperature of 100°C using an R F magneComposite films of RETM alloys and garnets have
been produced, with the objective of combining sensitivity
and storage density of amorphous compounds with the large
Kerr angles of the crystalline ferrites. (BiY),(FeGa),O, layers have been covered with DyFe[2371and TbFe.[2381The
Kerr angles of the composite layers were greater by a factor
of 10 compared to the values of the amorphous compounds.
5.3. Materials for Reversible Phase Change
Reversible phase change of inorganic materials is the second alternative to the magnet technology which is currently
under intensive investigation. Transitions between light-scattering crystalline and highly reflective amorphous phases
permit the switching. The crystalline materials-alloys
elements of the fifth and sixth main group-change into an
amorphous phase after pulsed laser irradiation from the melt
with rapid cooling. The crystalline phase is formed again by
extended heating beyond the glass point. The advantages of
this storage technique are the simple optical system of the
disk drive and high stability of the medium. Disadvantages
are the incomplete recrystallization in the event of frequent
overwriting, thus deforming the atomic lattice and destabilizing the inscribed information on repeated reading.['46. 2401
This can be reduced by matching the laser energy to the
number of cycles.[2411
The reversible properties of amorphous tellurium alloys
were observed for the first time by O v . ~ h i n s k y . Using
electron microscopy, it was found that amorphous and crystalline phases respectively are present which differ in their
reflectivity.[2431The rates of crystallization from the amorphous condition were investigated on various materials and
were ascribed to crystallization effects intensified by photons.12441Other authors have not confirmed this finding.[2461
In principle, the possibilities exist of writing either by the
change from amorphous to crystalline o r from crystalline to
amorphous. Von Gulfeld et al. were the first to propose the
phase change from crystalline to amorphous as a writing
mode[2451which is also referred to as RM (Reverse Mode)
The advantage of this mode of data marking is that the
crystallization rate limits the switching speed in the amorphous-crystalline mode. This is of importance with regard to
computer application with the shortest possible write times;
where erasure from amorphous to crystalline is concerned,
longer switching times can be tolerated. Figure 28 represents
the temperatures and pulse times for writing, reading and
erasure in the case of R M recording. Before being used, the
storage medium is heated by a floodlight (approximately
writing reading
Fig. 28. Principle of reverse mode recording with amorphous (A) to crystalline
(C) phase change materials [ 6 ] .The width of the signals corresponds to the pulse
1-10 ps), in order to crystallize the film for the starting condition. Using short pulses-approximately
100 nsec and
shorter-the marking of the amorphous domains takes place
at high laser power. In this case, the material is melted briefly
so that the crystalline arrangement is destroyed. After the
pulse, the medium cools down at cooling rates of lo910, K s- whereby the formation of a regular crystal lattice is prevented. The read beam has only a fraction of the
energy of the write beam, in order to permit the appearance
of changes even at very high numbers of cycles ( > lo6). The
energy of the erasure beam is selected so that the storage
medium is heated beyond the glass point and recrystallization commences.
The particular objective of current developments is to increase the speed of switching (crystallization) and the stability at room temperature in order to guarantee data security.
However, the improvement of the stability is also in most
cases associated with longer crystallization times and results
in longer pulse times for erasure. Within these conflicting
requirements there resides a fundamental difficulty regarding the development of phase-change materials which can be
used on a technical scale. The extension of the crystallization
can, for example, follow from a phase separation taking
place simultaneously during the crystallization since in this
case diffusion-controlled transport processes limit the speed
of crystallization.
Reference will be made herein to some current materials.
High linear bit densities of 0.63 pm per bit were achieved
with a PbTeSe alloy, which permitted, in a specific practical
form, an increase in the density to 0.50 pm per bit.[2471In this
way, it is possible to store approx. 1 GByte data capacities
per side on a 5.25" disk. Depending upon the tellurium content, indium-containing films of the In,, _xSb5,Te, type may
be switched very rapidly: with x = 1 , 2, 5, 30 and 20, a pulse
width of 50, 70, 90, 300 ns and 1 ps respectively was measured for the recrystallization from the amorphous
InSeTlCo alloys show a rapidly switchable phase change
from amorphous to crystalline: the phase transformation
takes place in 60 ns at a laser power of 14 mW[2491(Fig. 29).
Based on a crystallization temperature of 320 "C and an acti-
Angew. Chem. Inl. Ed. Engl. 28 (1989) 1445-1471
Fig. 29. Laser energy and pulse duration for the reversible phase transformation of a n InSeTlCo alloy at two film temperatures [249].
vation energy of 3 eV, the result is approximately 100 years
for the lifetime of amorphous domains. During recent times,
progress has been reported regarding GeTeSr~,[~~’l
GeTeTi.[2511and InSePb[2521alloys.
films,[2551demixing effects in polymers,[2561crystallization
effects in doped polymer m a t r i ~ e s , l ~specific
~ ’ ~ fluorescence
dyes,”581 luminescence in doped aluminum oxide,[2591or
storage processes utilizing the piezoelectric or pyroelectric
effects,[2601scanning tunneling microscopy,[z6‘I the birefringence relaxation of stretched dye-polymer films,[26z1
and the
LIESST effect (Light Induced Excited Spin State Trapping).[2631With other recording processes, the information
densities of the industrially produced optical memories may
even be exceeded: photochemical hole burning 10‘ bit
cm-2,[2641electron beam solid state memories 10”bit
cm-2[2651 and holographic memories lot4 b i t ~ m - ~ . [ ~ ’ ~ ’
These recording media and processes have no practical significance yet; however, they form part of intensive research
and development efforts as is evident from a research program of the Japanese MITI[2661and more recent publications in the field of photochemical hole burning.[2671
n u m b e r of a t o m s
6. Summary and Outlook
As a result of the success of CD technology, the development of write-once (WORM) and reversible (EDRAW) storage disks to the stage of marketing readiness has become a
matter of prime importance in industry during recent years.
The IR emission of the semiconductor lasers employed as
light source makes the synthesis of novel IR-absorbing dyes
necessary for WORM recording media. Specific compounds
of the category of the polymethines, phthalo- and naphthalocyanines and dithiolatonickel complexes have already
been used with success. In addition, metal alloys have been
recognized to permit storage by means of a phase change.
WORM memories are mainly used for data archiving; accordingly, a high degree of data security over long periods of
time is required. Due to the sensitivity to oxidation of the
inorganic WORM layers, there are some advantages of the
dye media.
The most highly developed EDRAW memories are represented by the magneto-optical films consisting of rare earth/
element/transition metal alloys. It is expected that they will
be able to replace the currently dominant mass memories on
the basis of magnet technology within broad fields. Their
advantages are relatively high storage density, insensitivity
to mechanical stress, and easy exchangeability and thus they
follow the development trends of computer technology.
Developments of new optical disk drives will be a great
incentive with regard to storage media. The use of optical
reading systems based on laser arrays, of integrated optical
systems together with polymeric optical waveguides, and of
lasers with frequency doubling to generate smaller wavelengths will open up possibilities for further increasing the
data rates and storage densities. In the near future, this will
also initiate the production of novel recording layers. Alternatives to the memories discussed above could be represented by storage media which are based on photochromic
dyes,[2531liquid-crystal polymers,[2541Langmuir-Blodgett
Anprw. C h i n . In,. Ed. Engl. 28 (1989) 1445 -1471
Fig. 30. Comparison for different storage technologies of number of atoms
used to store a bit; the line represents the development of future technologies
(after [2c]). H = Fixed disk, A = magnetic bubble memory, 7 = thin layer
technique, 0 = optical memory, 0 = molecular storage.
Projects which are concerned with the construction of
molecular switches and molecular storage elements are
framed in terms of molecular dimensions.[2681This work is
concerned with the construction and the characterization of
isolated simple models for investigating fundamental phenomena and the further development of model concepts concerning possible modes of operation of molecular memories.
In this connection, conceptual considerations take account of the significance of molecular order and supramolecular structures. Against this background, attempts have been
made to utilize phenomena of self-organization in order to
generate orientation and structure. Thus: there are parallels
with the biological information system, which is characterized by the formation of highly ordered molecular systems.
It appears that even future information systems can be developed using concepts which are based on principles realized in
We express our thanks to Dr. B. Fischer, Dr. B. Lohr, and Dr.
W. Wiedemann (Information Technology Division of Hoechst
AG, Wiesbaden) for he!vjul discussions, to Mrs. P. Sendelbeck, Mrs. S. Nink, Mrs. E. Himburg, Mrs. U. Jensen and
Mrs. U. Kletschke,for typing the article, and to Mrs. G. Wettlaufer and Mr. M . Utermark for the graphical material.
Received: June 21, 1989 [A 738 IE]
German version: Angew. Chem. 101 (1989) 1475
[I] H. Kuhn, IBM J. Res. Dev. 32 (1988) 37.
[2] 8) R. Ldndauer, Phys.1971 Plenurvortr. Physikerlag. 36th 1971 286;
h) H. Kuhn, Ber. Bunsenges. Phys. Chem. 80 (1976) 1209; c) R. W. Keyes.
I B M J. Res. Dev. 32 (1988) 24.
[3] G. Bouwhuis, J. Brdat, A. Huijser. J. Pasman, G. van Rosmalen, K.
Schouhamer-Immink: Principles of Optical Disc S.wtems, Adam Hilger,
Bristol. UK 1985.
[4] J. Isailovic: Videodisc and Optical Memory Sjstrms, Prentice-Hall, Englewood Cliffs, NJ. USA 1985.
[5] a) R. A. Sprague, (Ed.), Proc. SPIElnt. Soc. Opt. Eng. 329 (1982), b) D.
Chen (Ed.), ibid. 382 (1983); c) E. V. La Budde (Ed.). ihid 421 (1983);
d) R. A. Sprague (Ed.), ihrd. 529 (1985); e) D. A. Carlin, Y Tsunoda,
A. A. Jamberdino (Ed.). ihid. 899 (1988); f) G . R. Knight, C. N. Kurtz
(Ed.), ihid. 1078 (1989).
[6] A. E. Bell in M. J. Weber (Ed.): Handhook ?/Laser Scienceand TKhnohgj. Vol. 5 , Purr 3 , C R C Press, Boca Rota, FL, USA 1987, p. 65.
[7] a) H. H. Hopkins. J. Opt. Sot. Am. 69 (1979) 4; b) M. G. Carasso,
J. B. M. Peek. J. P. Sinjou, Philips Tech. Rev. 40 (1982) 151; c) J. P. J.
Heemskerk, K. A. Schouhamer-Immink. ihid. 40 (1982) 157; d) L. J.
Laub, IEEE Trans. Consum. Electron. CE-22 (1976) 258.
IS] a) G. C. Kenney. IEEE Trans. Consum. Electron. CE-22 (1976) 327; h) I.
Sebestyen. Electron. Pub. Rev. 2 (1982) 41; c) C. M. Goldstein, Science
( Washington) 215 (1982) 862; d) M. Noordenhos, Mikrodok. (FBOFachverlag; ISSN 0344-8010) 14 (1988) 8 ; e) R. P. Freeze, IEEE Spectrum 25 (1988) 41; f) D. Caruso. Electronics 58 (1985) 26.
[9] A. E. Bell, SID Proc. 24 (1983) 17.
[lo] M. H. Kryder, J. Appl. Phys. 57 (1985) 3913.
[ l l ] G. Kaempf, H. Loewer, M. W. Witman, Polym. Eng. Sci. 27(1987) 1421.
1121 F. E. Luborsky, Mater. Res. Soc. Symp. Proc. 80 (1987) 375.
1131 J. E. Kuder, J Imaging Technol. 12 (1986) 140.
1141 J. E. Kuder, J. Imaging Technol. 32 (1988) 51.
1151 J. M. Pearson. C R C Crit. Rev. Solid State Mater. Sri. 13 (1986) 1
[I61 D. S. Bloomberg, G. A. Neville Connell in C. D. Mee, E. D. Daniel (Ed.):
Magnetic Recording, Vol. 2 , McGraw-Hill, New York 1988; p. 304.
[17] M. Fujimoto, G. Sato, Shikizai Kyokai.shi 61 (1988) 215.
1181 F J. A. M. Greidanus, S. Klahn. Angew. Chem. Adv. Muter. 101 (1989)
243; Angew. Chem. Int. Ed. Engl. Adv. Muter. 28 (1989) 229; Adv. Mater.
1989, 45.
1191 a ) N. Grote. Phys. Unserer Zeit 8 (1977) 103; b) K. J. Ebeling, Laser
Gptoelekn. 16 (1984) 176; c>W. T. Tsang, Appl. Phys. L e f f .42(1983) 650;
d) C . H. Good: Injection Electroluminescent Devices, Wiley, New York
1201 a) A. Watabe, K. Katoh, H. Ukita, Proc. SPIE Int. Soc. Opt. Eng. 529
(1985) 150; b)O. Ito, K . Yoshimoto, I. Watanabe, T. Kubo, ibid. 529
(1985) 145; c) R. Miller, C. Abate, ibid. 529 (1985) 140; d) Y. Tsunoda, S.
Horigome, Z. Tsutsumi, ibid. 382(1983) 24; e) C. Bricot. J. Cornet, J.-L.
Gerad, F. Le Carvennec, ibid. 329 (1982) 94; f) I. Prikryl, ihid. 1078 (1989)
244; g) K. Katoh, A. Watabe, M. Yamamoto, ihid. 899 (1988) 24.
[21] M. Nagashima, Appl. Phys. Lett. 42 (1983) 144.
(221 a) A. T. Eventoff, G. C. Kenney, Proc. SPIE I n f . Soc. Opt. Eng. 329
(1982) 252; b) D. H. Davies, ihid. 420 (1983) 53; c) D. B. Kay, ihid. 382
(1983) 97.
1231 K. Ohta, A. Takahashi, T. Deguchi. T. Hyuga. S. Kobdyashi, H. Yamaoka, Proc. S P l E Int. Soc. Opt. Eng. 382 (1983) 252.
1241 R. N. Gardner, T. Rinehart, L. H. Johnson, R. P. Freese, R. A. Lund,
Proc. SPIE lnt. Soc. Opt. Eng. 420 (1983) 242.
1251 J. H. T. Pasman, H . F. Olijhoek, R. Verkaik, Proc. SPlE Int. Soc. Opt.
Eng. 529 (1985) 62.
[26] C . M. Rodid, Proc. SPIE In[. Soc. Opt. Eng. 529 (1985) 69.
1271 K. Manabe, M. Niwano, M. Isohe, S. Shimokihara, Polym. Prepr. ( A m .
Chem. Soc. Div. Polym. Chem.) 29 (1988) 230.
128) P. E. J. Legierse. Galvanorechnik 79 (1988) 3246.
[29] J. G. Dil, C. A. Weshop, Appl. Opt. 18 (1979) 3198.
[30] Y. Okino, K. Sano, T. Kashihara, Proc. SPIE Inr. Soc. Opt. Eng. 329
(1982) 236.
1311 M. J. Brekner, Pol.vm. Prepr. ( A m . Chem. Soc. Div. Polym. Chem.) 29
(1988) 234.
1321 J. Hennig, Jpn. J. Appl. Phys. 26, Suppl. 24-4 (1987) 9.
1331 J. Hennig, KunststoJfe 75 (1985) 425.
1341 M. Nakada, M. Saitoh, K. Toki, H. Gokan, Jpn. J. Appl. Phys. 26. Suppl.
24-4 (1987) 95.
1351 N. A. Peppas. C. N. Bowman. A. L. Carver, S. L. Kennett, M. M. Wilhams, Polym. P r c y . ( A m . Chem. Soc. Drv. Po/ym Chem.) 29 (1988)
1361 W. G. Driscoll (Ed.): Handbook of Optics, McGraw-Hill, New York 1978.
1371 A. Nussbaum. R. A. Philips: Comtemporury Optics for Scientisis and
Engineers, Prentice Hall, Englewood Cliffs, NJ, USA 1976.
1381 B. R. Hahn, J. H . Wendorff, Po/vmer 26 (1985) 1619.
[39] M. Isohe, S. Imai, Sumitomo Chem. Co., Ltd., Ehrme Res. Lab. 1986, 79
1401 K.Compaan, P. Kramer, Philips Tech. Rev. 33 (1973) 178.
1411 R. G. Gossink, Angew. Mukromol. Chem. 1451146 (1986) 365.
1421 H. C. Haverkorn, P. E. J. Legierse, G. E. Thomas, Philips Trch. Ree. 40
(1982) 287.
1431 J. G. Kloosterhoer. G. J. M. Lippits. H. C. Meinders, Philips Tech. Re!'.
40 (1982) 298.
1441 H. VO1z. Nuchrichtentech. Elektron. 37 (1987) 164.
[45] J. J. Wrobel. A. B. Marchant. D. G. Howe. Appl. Phys. Lett. 40 (1982)
1461 D. Meyerhofer, J. Appl. Phys. 49 (1978) 3993.
1471 a) V. B. Jipson. C. R. Jones, J. Vac. S o . Technol. 18 (1981) 105; b) A. E.
Bell, F. W. Spong, IEEE J. Quantum Electron. QE14 (1978) 487.
[481 K. Y. Law. P. S. Vincett. R. 0 . Loufty, L. Alexandru, M. A. Hopper,
J. H. Sharp. G. E. Johnson, Appl. Phj-s. Lett. 36 (1980) 884.
1491 a) K. Y. Law, P. S. Vincett, G. E. Johnson. Appl. Phys. Lett. 39 (1981)
718; b) D. G. Howe. J. J. Wrobel, J. Vuc. Sci. Technol. 18 (1981) 92.
[501 a) T. W. Smith, D . Wychik, J. Ph,vs. Chem. 84 (1980) 1621; b) J. Drexler,
J. Vuc. Sri. Technol. 18 (1981) 87.
1511 M. C. Guptd, Appl. Opt. 23 (1984) 3950. .
[52] M. Mitsuya, M. Terao. T. Kaku, K. Shigematsu, M. Akagi, J. Appl. Phys.
54 (1983) 3710.
(531 a) V. Novotny, L. Alexandru, J. Appl. Phw. 50 (1979) 1215; b) J. Appl.
Poljm. Sci. 24 (1979) 1321.
1541 a) L. Alexandru, M. A. Hopper, R. 0 .Loufty, J. H. Sharp, P. S. Vincett,
G. E. Johnson. K. Y. Law, Polvm. Prepr. ( A m . Chem. Soc. Dir.. Polym.
Chem.) 25 (1984) 305; b) L. Alexandru, M. A. Hopper, R. 0. Loufty,
J. H. Sharp. P. S. Vincett. G. E. Johnson, K. Y. Law m A. Davidson
(Ed.): Muteriuls,for Micrulithographj ( A C S Symp. Ser. 219 (1983)).
1551 a) J. A. Cornet, Proc. SPIE Int. Soc. Opt. Eng. 420 (1983) 86; h) USPat. 4360895 (1982), Thomson C S F S.A.; c) J. Cornet, J. C. Lehureau.
H. Magna. EP-A 33046 (1981). Thomson CSF S.A.; d) K. N. Mafitt,
W B. Robbins. R. F. Williams. US-Pat. 4430659 (1984), Minnesota Mining & Mfg. Comp.
1561 D. Mayden, B e N S w . Tech. J. 50 (1971) 1761.
1571 P. Kivits, R. de Bont, Appl. Phys. 24 (1981) 307.
[SSl a) P. Kivits, R. de Bont. P. Zaim, Appl. Phys. 24 (1981) 273; b) P. Kivits,
R. de Bont, B. Jacobs, P. Zalm, Thin Solid Films 87 (1982) 215.
[59] T. S. Chung, J. Appl. Phys. 60 (1986) 55.
[60] K. Y Law, G. E. Johnson, J. Appl. Phys. 54 (1983) 4799.
1611 M. F. Molaire: The Influence of Melt Viscosit.v on the Writing Sensitivity
of Organic Dye-Binder Optical Disk Recording Media ( O S A Top. Meet.
Opt. Data Storage, Stageline, N K U S A 1987, Abstr. ThA4-1.
[62] a) A. Bloom, R. A. Bartolini, A. E. Bell. US-Pat. 4023 185 (1977). RCA
Corp.; h) H. T. Thomas, J. J. Wrobel, US-Pat. 4380769 (1983) Eastman
Kodak Comp.; c) S. R. Gerfast, US-Pat. 4 365 257 (1982), Minnesota Mining& Mfg. Comp.; d ) B. Albert, P. Neumann, G. Hansen, K. D. Schomann, DOS 34 28 458 (1986), BASF AG.
1631 A. E. Bell. R. A. Bartolini, US-Pat. 4 195312 (1980), RCA Corp.
1641 A. Bloom. R. A. Bartolinl, US-Pat. 4218689 (1980) RCA Corp.
[65] A. Kuroiwa. K. Namba, S. Asami, T. Aoi. K. Takahashi, S. Nakagawa.
Jpn. J. Appl. Phys. Part 1 22 (1983) 340.
[66] M. Kunikane. Jpn. Kokai Tokkyo J P 58-1 12 793 (1983). Ricoh Co., Ltd.;
Chem. Absrr. 99 (1983) 167101.
[67] Jpn. Kokai Tokkyo Koho J P 56-46221 (1981), Nippon Telegraph and
Telephone Public Corp.; Chem. Ahstr. 95 (1981) 178630.
1681 Jpn. Kokai Tokkyo Koho J P 58-1 12 790 (1983). Ricoh Co., Ltd.; Chem.
Ahstr. 99 (1983) 167102.
1691 M. Kunikane, M. Ahe, Jpn. Kokai Tokkyo JP58-I12791 (1983). Ricoh
Co., Ltd.; Chem. Abstr. 99 (1983) 131425.
1701 P. Kivits, R. de Bont, J. van der Veen, Appl. Phyx A26 (1981) 101.
[71] a) M. Kunikane, M. Abe, M. Umehara, US-Pat. 4460665 (1984), Ricoh
Co., Ltd.; b) K. Namha, S. Asami, A. Kuroiwa, A. Nakagawa. Jpn.
Kokai Tokkyo J P 59-55 795 (1984). TDK Corp.; Chem. Abstr. 102 (1985)
87717; c ) Jpn. Kokai Tokkyo JP 59-85791 (19841, Ricoh Co., Ltd.;
Chem. Abstr. 101 (1984) 181296; d) Jpn. Kokai Tokkyo J P 59-24690
(1984). TDK Corp.; Chem. Ahstr. 100 (1984) 2191 14; e) Jpn. Kokai Tokkyo J P 59-67092 (1984), TDK Corp.; Chem. Abstr. 102 (1985) 36849;
O H . Oba. Y. Ueda, T. Sato, M. Umehara, M. Abe, H. Funakoshi, S.
Kobayashi, GB-Appl. 2 165658 (1986), Ricoh Co., Ltd.
[72] a) E. Daltrozzo, G. Scheibe, K. Gschwind, F. Haimerl, Photogr. Sci. Eng.
18 (1974) 441, b) T. M. Herz. ibid. 18 (1974) 323.
1731 G . W. Byers. S. Gross, P. M. Henrichs, Photochem. Photohiol. 23 (1976)
1741 H. Oba, T. Sato, M. Umehara. M. Abe, Jpn. Kokai Tokkyo JP 61-248 789
(1986), Ricoh Co., Ltd.; Chenr. Abstr. 107 (1987) 49658.
Angew. Chem. Im. Ed. Engl. 28 (1989) 1445-1471
[75] a) Jpn. Kokai Tokkyo JP58-195595 (1983), Canon KK; Chem. Abstr.
101 (1984) 238217; b) Jpn. Kokai Tokkyo JP 58-220792 (1983). Canon
KK:Chem. Ahstr. 101(1984)81726:c)Jpn. KokaiTokkyoJP 58-219091
(19x3). Canon K K ; Chem. Absrr. 101 (1984) 81727: d)Jpn. Kokai Tokkyo JP 59-150795 (1984) Ricoh Co., Ltd.: Chem. Ahstr. 102 (1985)
[76] a) K . Miurd, T. Ozawa, J. iwanami, Jpn. Kokai Tokkyo JP 60-228448
(1986). Mitsubishi Chemical Industries Chem. Ahstr. 104 (1986) 129641 :
b) Jpn. Kokai Tokkyo JP 61-167 680 (1986), Mitsubishi Chemical Industries; Chem. Ahstr. 106 (1987) 50042; c)Jpn. Kokai Tokkyo JP61167 681 (1986). Mitsubishi Chemical Industries; Chem. Abstr. 106 (1986)
50044: d) T. Ozawa. S. Maeda, Y. Kurose, Jpn. Kokai Tokkyo JP 62138459 (1987). Mitsubishi Chemical Industries: Chrm. Ahstr. 107 (1987)
[77] a) Jpn. Kokai Tokkyo J P 58-217 391 (1983). Asahi Chemical Industry
Co., Ltd.: Chem. Abstr. 102 (1985) 36847: b) Jpn. Kokai Tokkyo JP 58173696(1983).Canon KK:Chem. Abstr. 100(1984)94591;c) Jpn.Kokai
Tokkyo JP 59-14150(1984), Canon KK; Chem. Ahstr. 102(1985) 15216.
1781 a ) W. Schrott, B. Albert, P. Neumann, H. Benthack-Thoms, DE0s 3703985 (1988), BASF AG; b) T. Hioki, Y Inagaki, M. Yabe, DE0 s 38 06 181 (1988). Fuji Photo Film Co., Ltd.; c) T. Ozawa, S. Maeda.
Y. Kurose. Jpn. Kokai Tokkyo JP 62-130890 (1987). Mitsubishi Chemical Industries: Chem. Ahstr. 107 (1987) 226074.
1791 a) M. Dust, P. Neumann, DE-OS 3622 138 (1988), BASF AG: b) Y.Ma,
S. Ishikawa, T. Arita. K. Sakai, M. Maikara. Jpn. Kokai Tokkyo J P 5984249 (1984). Canon KK; Chem. Ahstr. 102 (1985) 103493.
[SO] K. Sakai, S. Ishikawa, T. Arita. M. Mabuchi, S. Umehara. USPat. 4501 808 (1985). Canon KK.
[81] a) M. Dust, P. Neumann, E. Schefczik. H. Benthack-Thoms, H. Barzynski. K. D. Schomann. H. Kuppelmaier, E. Koster. DE-OS 3608214
(19x7). BASF AG; b) M. Dust, P. Neumann, P. Hauser, G. Wagenblast,
H. Benthack-Thoms, H. Barzynski, K. D. Schomann, H. Kuppelmaier,
E. Koster. DE-OS 3631 843 (1988), BASF AG.
182) a ) G. L. Fletcher. S. L. Bender, D. H. Wadsworth, US-Pat. 4 577 024
(1986). Eastman Kodak Comp.; b) D. H. Wadsworth, H. T. Thomas,
G. L. Fletcher, C. H. Weidner, US-Pat. 4446223 (1984). Eastman Kodak
Comp.: c ) M. F. Molaire, US-Pat. 4499 165 (1984). Eastman Kodak
[83] M. A. Elmasry, WO 86/05504 (1986), Minnesota Mining& Mfg. Comp.
[84] a) M. Thomas, P. Neumann, DE-OS 34 13 940 (1985). BASF AG: b) H.
Berneth, R. Raue, DE-OS 3704263 (1988), Bayer AG; c) Jpn. Kokai
Tokkyo JP 58-219 090(1983), Canon KK;Chem. Absrr. 101(1984)81728;
d) D. J. Gravesteijn, J. van der Veen, Philips. Tech. Rev. 41 (1983/1984)
325: e) D. J. Gravesteijn, C. Steenbergen, J. van der Veen, Proc. S P I E Int.
Soc. Opt. Eng. 420 (1983) 327.
[85] a) K. Namba, Jpn. Kokai Tokkyo JP 60-159087 (1985), TDK Corp.;
Chem. Ahstr 104 (1986) 120089; b) Jpn. Kokai Tokkyo JP 60-163243
(1985), TDK Corp.; Chem. Abstr. 106 (1987) 11249; c) Jpn. Kokai Tokkyo Koho JP 59-55795 (1984), TDK Corp.; Chem. Abstr. 102 (1985)
87717; d) Jpn. Kokai Tokkyo Koho J P 59-67092 (1984), TDK Corp.;
Chrm. Ahstr. 102 (1985) 36849.
1861 a)Jpn. Kokai Tokkyo Koho JP59-55794 (1984). TDK Corp.; Chem.
Ahstr. 102 (1985) 70319; b) Jpn. Kokai Tokkyo Koho JP 59-81 194
(1984). TDK Corp.; Chem. Absrr. 102 (1985) 140964; c) S. Yamamuro,
M. Abe, Y Ueda, T. Sato, M. Umehara, Jpn. Kokai Tokkyo Koho JP 60257290 (1985) Ricoh Co., Ltd.; Chem. Abstr. 10s (1986) 15357; d) Jpn.
Kokai Tokkyo Koho JP 59-83695 (1984), TDK Corp.; Chem. Ahstr. 102
(1985) 70323.
I871 a ) P. V. Susi. DE-OS 1816247 (19691, American Cyanamid Corp.;
b) P. V. Susi, N. A. Weston, FR-Appl. 1398 240 (1965), American Cyanamid Corp.; c) T. Salo, M. Umehara, M. Abe, H. Oba, Y. Ueda, GBAppl. 2 155 811 (1985). Ricoh Co., Ltd.; d) T. Sato, S . Shindo, T. Numa,
M. Sumiya. Jpn. Kokai Tokkyo Koho JP 62-32 132 (1987), Nippon Kayaku Co., Ltd.; Ckem. Ahstr. 107(1987)24277; e) T. Sato, M. Umehara, M.
Abe. H. Oba, Y. Ueda, DE-OS 3503995 (1988), Ricoh Co., Ltd.
[88] a) K . Namba, EP-A 147083 (1985), TDK Corp.; b) K. Namba. S.Asami, T Aoi, K. Takahashi, A. Kuroiwa, Jpn. Kokai Tokkyo Koho JP 618 384 (1986). TDK Corp.; Chem. Abstr. 105 (1986) 70213 ; c) S. Yamamuro. H. Oba. M. Abe, T. Sato, Y Ueda, M. Umehara, Jpn. Kokai Tokkyo
Koho J P 61-72586 (1986). Ricoh Co., Ltd.; Chem. Ahstr. 106 (1987)
1891 a) T. Sato. T. Sakaeda, K. Ichinose, Jpn. Kokai Tokkyo Koho JP 62193891 (1987), Ricoh Co., Ltd.; Chem. Abstr. 109 (1988) 160675; b)T.
Sato. DE-PS 3521 915 (1985). Ricoh Co., Ltd.; c) K. Namba. I"' lnt.
S.vmp. Chem. Funct. Dyes, Kinki Chemical Society, Osaka, Japan 1989;
d) H. Flohr, J. Jesse, B. Albert, P. Neumann, DE-OS 35 20 109 (1986),
BASF AG: e)T. Sato, T. Eida, K. Ichinose, DE-OS 3724981 (1988)
Ricoh Co., Ltd.
1901 Neth. Appl. NL 83 00 155 (1984), Philips Gloeilampenfabrieken.
I911 a) F. H. Moser, A. L. Thomas: The Phthalocyanines, Vol. I. 11, CRC
Press, Boca Rota, FL, USA 1983; b) A. B. P. Lever. Adv. lnorg. Chem.
Rudiochem. 7 (1965) 27; c) D. Wohrle, G. Meyer, Kontakte (Durmstudt)
I Y X S . No. 3. p. 38.
Annew. C'hcm. Inr. Ed. En$
28 (1989) 1445-1471
[92] a) F. H. Moser. A. L. Thomas: Phthalocymine Compound.\. Reinhold,
New York; Chapman and Hall, London 1968: b) L. J. Boucher in G. A.
Meslon (Ed): Coordination Chemistry o / Macrocjclic Compound.s, Plenum, New York 1979; Chapter 7.
[93] a) A. R. Kane, J. F. Sullivan. D. H. Kenney, M. E. Kenney, lnorg. Chem.
Y (1970) 1445; b) B. L. Wheeler, G. Nagasubramanian. A. J. Bard. L. E.
Schechtman, D. R. Dininny. M. E. Kenney, J. Am. Chrm. Soc. 106 (1984)
[94] a) M. Hanack, J. Metz, G. Pawlowski. Chem. Ber. 11s (1982) 2836; b) D.
Wohrle, G. Meyer, B. Wahl, Makromol. Chem. 181 (1980) 2127; c ) C.
Piechocki, J. Simon, A. Skoulios, D. Guillon, P. Weber, J. Am Chem. Six.
104 (1982) 5245.
[95] a) S. Higaki, K. Hanabusa, H. Shirai, N. Hojo. Makromul. Chem. 183
(1984) 691: b) G. Pawlowski, M. Hanack, Synthesfs 1980. 287.
[96] Y. Goto, N. Kohino, H. Goto, K. Ogawa. EP-A 83991 (1983). Fujitau
I971 a) A. M. Hor, R. 0. Loufty, Thin Solid Frlms 106 (1983) 291: b) R. 0.
Loufty, A. M. Hor. G. DiaPaola, C . K. Hsiao. J. Imuxinn Sf.29 (1984)
[98] Jpn. Kokai Tokkyo Koho JP 59-16785 (1984). Surnitomo Chemical Co..
Ltd.; Chem. Abstr. 100 (1984) 148595.
1991 a) K. Tanikawa, US-Pat. 4,458,004 (1984. Ricoh Co.. Ltd.); b) 0. Kaieda. I. Okitaka, H. Ito. Jpn. Kokai Tokkyo Koho JP63-30566 (1988).
Nippon Shokubdi KK Co., Ltd.; Chem. Abstr. 109 (1988) 75212: c) 1.
Okitaka, H. Ito, 0. Kaieda, Jpn. Kokai Tokkyo Koho J P 63-141 342
(1988), Nippon Shukobai KK Co., Ltd.; Chem. Ahstr. flO(1988) 77513.
[loo] a ) A. Kuroiwa, N. Namba, T. Kamijo, Jpn. Kokai Tokkyo Koho JP 61152685 (1986). TDK Corp.; Chem. Ahsrr. 106 (1987) 20000: b)Jpn.
KokaiTokkyo Koho JP 61-246091 (1986),TDKCorp.; Chem. Ahsrr. 106
(1987) 224577; c) S . Ehashi, S. Miyazaki, M. Sakamoto, Jpn. Kokai Tokkyo Koho JP 63-37991 (1988). Toyo Ing. Mfg. Co., Ltd.; Chem. Ahstr.
109 (1988) 119826; d) M. Sakamota, S. Miyazaki, T. Sato. S. Ehashi, 1"'
Int. Symp. Chem. Funct. Dyes, Kinki Chemical Society, Osaka. Japan
1989; e) H. Shibano, H. Yamazaki, K. Morikawd, Jpn. Kokai Tokkyo
Koho JP 63-39 388 (1988) Kao Corp.; Chem. Ahstr. 109 (1988) 46307.
[I011 a) S. A. Mikhalenko, E. A. Lukyanets, Z h . Ohshch. Khim. 39 (1969)
2129; b) H. Itoh, S. Takahara. M. Matsumoto, Is' h i t . Simp. Chem.
Funct. D.7e.s. Kinki Chemical Society, Osaka. Japan 1989.
[lo21 S. A. Mikhalenko, E. A. Lukyanets, Z h . Ohshch. Khim. 39 (1969) 2554.
[103] E. I. Kovshev. E. A. Lukyanets, Z h . Obshch. Khim. 42 (19721 696.
I1041 a) K. Y. Law, G . E. Johnson, J. W. P. Lin, US-Pat. 4492750 (1985), Xerox Corp.; b) S. Hirose, K. Abe, H. Ozawa, Y Hosono, EP-A 188 331
(1986), Mitsui Toatsu Chemicals, Inc.
[lo51 s. Tai, S. Hayashida, N. Hayashi. Y. Iwakabe, S. Numatd. N. Kinjo. S.
Era, S. Kobayashi, A. Mukoh, Y Sato. EP-A 284 370 (1988). Hitachi
Chemical Co., Ltd./Hitachi, Ltd.
I1061 S. Maeda, Y. Kurose. Y Wada, EP-A 279 426 (1988). Mitsubishi Kasei
[lo71 a) S. Hayashida, S. Tai, EP-A 254553 (1988), Hitachi Ltd./Hitdchi Chemical Ltd. Co., Ltd. b) R. S. Jones, D. E. Nikles, M. E. Kenney. EPA 279 426 (1986), Celanese Corp.; c) T. Seiji, S. Hayashida. H. Hagiwara,
M. Katayose, K. Kamijima, T. Akimoto, S. Era, S. Kobayashi. A. Mukoh. EP-A 296876 (1988), Hitachi Chemical Co., Ltd.
[lo81 T. Kanno, H. Watanabe, S. Nozaki, P r o f . 34'* Con/. Soc. Appl. Ph7.s.
1987, 720.
I1091 a) K. Morikawa, H. Shibano, H. Yamazaki, Jpn. Kokai Tokkyo Koho
JP 62-223287 (1987). Kao Corp.; Chem. Abstr. 109 (1988) 46307: b) S.
Hirose, H. Ozawa, K. Abe, Y. Hosono, Jpn. Kokai Tokkyo Koho JP 6357287 (IYSX), Mitsui Toatsu Chemicals, Inc.; Chem. Abstr. 109 (1988)
83689; c) S. Era, S. Kobayashi, A. Mukoh, EP-A 279 501 (1988), Hitachi
[110] B. Albert, H. Kuppelmaier, G. Wagenblast, DE-A 37 11 762 (1988).
[ l l l ] a) E. Merian, Chimiu 13 (1959) 181; b) Am. Dyest. Rep. 48 (1959) 31.
11121 a) K. Y Chen, J. Griffiths, J. Chem. Res. Miniprinr 197X. 2319; b) K.
Takagi. M. Matsuoka, Y. Kubo, T. Kitao, Dyes P i p . 6 (1985) 75.
[I131 a) M. Itoh, S. Esho, K. Nakagawa, M. Matsuoka, Proc. S P I E lnt. Soc.
Opr. Eng. 420 (1983) 332; b) K. Takagi. M. Kawabe. M. Matsuoka, T.
Kitao, Dyes Pigm. 6 (1985) 177; c) S. H. Kim, M. Matsuoka, Y Kubo, T.
Yoshida, T. Kitao, ibid. 7 (1986) 93; d) S. H. Kim, S. Minami. M. Matsuoka, T. Kitao, Chem. Express 2 (1987) 73.
11141 a) Y Kogo, H. Kikuchi, M. Matsuoka, T. Kitao, J. Soc. Dvers Colour. 96
(1980) 475; b) M. Matsuoka, K. Takagi, H. Obayashi, K. Wakasugi, T.
Kitao. ihid. 99 (1983) 257.
11151 a)Jpn. Kokai Tokkyo Koho JP58-112793 (1983). Ricoh Co., Ltd.:
Chem. Abstr. 9911983) 167101; b) Jpn. Kokai TokkyoKohoJP 59-76296
(1984), NEC Corp.; Chem. Abstr. 102 (1985) 103688; c) Jpn. Kokai Tokkyo Koho JP 59-76297 (1984), NEC Corp.; Chem. Ahstr. 102 (1985)
140967; d) S. Esho, M. Itoh, M. Matsuoka, Jpn. Kokai Tokkyo Koho
JP 59-5096 (1984) TDK Corp.: Chem. Abstr. 100 (1984) 148588.
[116] S. H. Kim, M. Matusoka, T. Kitao, Chem. Leti. 19x5. 1351.
El171 a) S . Edokoro, M. Itoh. M. Matsuoka, Jpn. Kokai Tokkyo Koho JP 6122-4588 (1986). NEC Corp.; Ckem. Abstr. I07 (1987) 49663; b) Jpn.
KokaiTokkyoKohoJP61-246092(1986).NECCorp.; C h ~ mAhstr.
(1987) 49665.
[I181 S. H. Kim, M. Matsuoka, T. Yodoshi. T. Kitao, Chem. / (1986)
[119] a)Jpn. Kokai Tokkyo Koho JP58-169152 (1983). Ricoh Co.. Ltd.;
Chem. Ahstr. 106 (1987) 147206; h) Jpn. Kokai Tokkyo Koho J P 58224448 (1984). T D K Corp.; Chem. Ahstr 101 (1984) 238255; c) W.
Schrott, P. Neumann. B. Wolff. A. Stange. DE-OS 3726458 (1989).
[120] a) I . L. Boguslavskaya, V. I. Gudzenko. J. Org. Chem. USSR 4 (1968) 96;
b) H. Shibano. K. Morikawd, H. Yamazaki. Jpn. KOkdi Tokkyo J P 6281 082 (1988), Kao Corp.; Chem. Ahstr. 109 (1988) 160749.
[I211 a ) M. J. Baker-Hawkes, E. Billig, H. B. Gray. J. Am. Chem. Soc. M(1966)
4870; h) H. Shiozaki. H. Kakazumi. Y. Nakado. T. Kitao. Chem. Lrtt.
1987, 2393.
11221 a) M. Suzuki, M. Kawdmura. K . Kato, M. Takahashi. T. Morishita, K.
Nakayama, A. Nakatsuka. EP-A 278 257 (1987). Seitetsu Kagdku Co..
Ltd.. h)W. Schrott, P. Neumann. B. Albert. EP 193774 (1986) BASF
AG;c) W. Schrott, P. Neumann, B. Albert. M. Thomas. H. Bdrzynski. K.
Schomann, H. Kuppelmaier, DE-OS 3505751 (1986). BASF AG; d) K.
Sasagawa. E. Noda. M. Imdi. WO 831’02428 (1983). Mitsui Todtsu Chemicals, Inc.
[123] G. N. Schrduzer, V. P. Mayweg, J. Am. Chern. Soc. 87 (1965) 1483.
11241 Y. Suzuki, G. Hayasht, DE-OS 3726013 (1988). Fuji Photo Film Co..
[I251 S. H. Kim, M. Matsuoka. M. Yomoto. Y. Tsuchiya, T. Kitao. Dycs P i p .
X (1987) 381.
[126] Y. Kuho. K. Sdsaki, K. Yoshida, Chem. Lm. 19x7, 1563.
[127] K. Yoshida, M. Ishiguro, Y. Kuho, Chem. Lert. 1987. 2057.
[128] H. Junek. G. Uray. G. Zuschnig, Dyes Pigrn. Y (1988) 137.
[129] a) S. Akiyama, S. Nakatsuji, K. Nakashima, M. Watanabe, J. Chem. Soc.
Chem. Commun. 1Y87,710,1240;h) S. Akiyama, S. Nakatsuji, K. Nakashima, S. Yamasaki, D.ves Pigm. 9 (1988) 459.
[130] a) R. West. D. C. Zecher, J. Am. Chem. Sac. 92 (1970) 155: h) W Schrott,
P. Neumann, DE-OS 3507 379 (1986), BASF AG.
I1311 H. Barzynski, H. Naarmann. EP-A 230240 (1987). BASF AG.
[132] M. Matsuoka, T. Yodoshi, L. Han, T. Kitao. Dyes Pigm. 9 (1988) 343.
[133] G. KHmpf, Ber. Bunsenges. Ph.ys. Chem. 89 (1985) 1179.
1134) A. E. Bell, Proc. SPIE Int. Sol. Opt. Eng. 382 (1983) 2.
[135] A. S. Haogland, Proc. SPIE Int. Soc. Opt. Eng. 382 (1983) 16.
[136] S. Yoshii. 0 .Oshii, S. Hattori. T. Nagakawa, G. Ishida, J. Appl. P h w 53
(1982) 2256.
[137] G. J Akashr, J. Telev. Eng. 40 (1986) 466
[I381 D. C. Hsieh, E. V. La Budde, R. J. Johnson, Proc. S P l E Int. Soc. Opt.
Eng. 329 (1982) 814.
[139] E. W. Pugh, D. L. Critchlow, R. A. Henle, L. S. Russell, IBM J. Rex Drv.
25 (1981) 585.
11401 J. M. Harker, D. W. Brede, R. E. Pattison, G. R. Sdntana, S. G. Tdft.
I B M J Res. Dev. 25 (1981) 677.
[141] a) A. Kunimoto, M. Kato, T. Masuda, S. Ohmuma, T. Masumoto. lEEE
Trans. Mugn. MAG-23 (1987) 2803; b) T. Osaka. N. Kasai, I. Koiwa, F.
Goto. Y. Suganuma, J. Electrochem. Soc. 130 (1983) 568.
[142] J. P. van der Ziel, R. A. Logan, R. M. Mikulyak, Appl. Phys. Lett. 41
(1982) 9.
[143] D. Boetz. J. C. Connolly, C. B. Gilbert, M. G. Harvey. M. Ettenberg,
Appl. Phys. Lett. 41 (1982) 1040.
11441 J. McLeod. Elrctronics 10 (1988) 70.
[145] M. L. Levene, Proc. SPIE In(. Soc. Opr. Eng. 899 (1988) 279.
[146] M. Ojima, N. Ohta, Hilachi Rev. 37 (1988) 139.
11471 F. Tandka. Jpn. J. Appl. Phys. 26 (1987) 231.
(1481 T. Nakao, Jpn. J. Appl. Phy.7. 26. Suppl. 26-4 (1987) 149.
[149] I. Koiwd. M. Toda, T. Osaka. J. Electrochem. Soc. 133 (1986) 597.
[150] T. Niihara, N. Ohta, K. Kaneko, Y. Sugita, S. Horigome, IEEE Trans.
Mugn. MAG-22 (1986) 1215.
[151] M. Asano, M. Kohayashi, Y Maeno, K. Oishi. K. Kawamura, lEEE
Trans. Mugn. MAG-23 (1987) 2620.
[152] H. Oda, S. Ozaki, Jpn. J. Appl. Phys. 22 (1983) 1930.
I1531 T. K. Hatwar. S. C. Shin, IEEE Truns. Mugn. MAG-22 (1986) 946.
[154] a) J. W. M. Biesterhos. A. G. Dirks, M. A. P. J. Fdrld, P. J. Grundy. Tl7in
Solid Films 58 (1979) 259; h) R. Hasegawd, R. C. Tylor. J Appl. Phys. 46
(1975) 3606; c ) Y. Mimura, N. Jamamura. T. Kohayashi. IEEE Truns.
Mugn. MAG-12 (1976) 779; d) J. M. E. Harper. R. J. Gamhino. J. Vuc.
Sci. Techno/. 16 (1979) 1901 ; e) R. B. van Dover, J. Appl. Phys. 59 (1986)
[155] H. Heitmdnn, M. Hartmann, M. Rosenkranz, H. J. Tolle, J. PIiys. Colloy.
1985, 9.
[156] F. E. F. E. Luborsky. Tshi Liuo K b Hsurh /Y (1987) 1987.
[157] K. Arita (Ed.): Development of’Opticu1 Recording Disk in DuicelChemicul
Industries, Daicel Chemical Industries, Tokyo 1987.
[lSX] H. J. Williams, R. C. Sherwood. R. F. G. Foster. E. M. Kelleg. J. Appl.
P h m 2X (1957) 1181.
11591 L. Mayer. J. Appl. Phvs. 29 (1958) 1003; 1454.
[160] J. T.Chang, J. F. Dillon, M F. Cianola. J. Appl. Phjs. 36 (1965) 1100.
[161] a) D. Chen, W. E. Stutius. IEEE Truns. Mugn. MAG-10 (1974) 581 ; b) S.
Esho. S. Noguchi, Y. Oho, M. Nagao, Appl. Opt. 13 (1974) 179: c) D.
Chen. G. N. Otto, F. M. Schmidt, IEEE Truns. Mug% MAG-9 (1973) 66;
d) D. Chen. Appl. Opt. 13 (1974) 767: e) H. E. Haskal, G. Bernal. D.
Chen. ibid. 13 (1974) 866: 0 S. Honda. T. Nomura. T. Kusuda. IEEE
Truns. Mugn. M A G 9 (1973) 467.
[162] W. K. Unger. E. Wolfgang, H. Harms, H . Handek, J. Appl. Phss. 43
(1972) 2875.
[163] a ) P. Chaudhari. J. J. Cuomo, R. J. Gamhino. Appl. Phys. Lett. 22 (1973)
337: h) IBM J. Rrs. Dei,. 17 (1973) 66.
[164] S. Matsushitd, K. Sundgo, J p n . J. Appl. Phys. 14 (1975) 1851.
[165] Y. Togami, K. Kobayashi. M. Kajiurd, K. Sato, T. Teranishi. Proc. S P l E
In!. Soc. Opt. Eng. 329 (1982) 208.
I1661 Y. Togami, K. Kobdyashi. M. Kajiura. K. Sato. T. Teranishi, J. Appl.
Phys. 53 (1982) 2335.
[167] Y. Togami, Appl. Phys. Lett. 32 (1978) 673.
[IbX] S. Honda, K. Meda. T. Kmuda. J. Appl. Phy.r. 52 (1981) 2295.
[169] S. Tatsuoke, Y. Togdmi, K i m Zuiryo 3 (1983) 9.
[170] Y. Kobayashi, Y. Togami, T. Teranishi, Jpn. J. Appl. Phw. 19 (1980) 581.
[171] Y. Togami. K. Kobdyashi. Jpn. J Appl. Phw. 19 (1980) 1457.
[I721 Y. Togami, K. Kobdydshi. 1. Yanagida, Jpn. J. Appl. Phy,s. / 9 (1980) 1663.
[173] Y. Hoshi. M. Naoe, S. Ydmanaka, J. Appl. Phys. 53 (1982) 2344.
[174] M. Ohkoshi, 0. Ryoji. K. Inoue, S. Honda. K. Tetsuzo. Jpn. J. Appl.
Ph.vs. I 9 (1980) 1807.
[175] S . Yasugi. S. Honda. M. Ohkoshi, T. Kuzuda. J. Appl. Phys. 52 (1981)
[176] A. G. Dirks. H. J. Leamy,J. Appl. Phys. 49 (1978) 1735.
[177] T. Katayama, M. Hirano. Y. Koizumi. K. Kawdnishi. IEEE Truns. Mugn.
MAG-13 (1977) 1603.
[178] G. A. N. Conell. R. Allen. J. Mugn. Mugn. Mutrr. 31-34 (1983) 1516.
[179] P. Hansen. J. Appl. Phys. 63 (1988) 2364.
[180] S. Tsunashima. S. Masui, T. Kobdydshi, S. Uchiydma, J. Appl. Phys. 53
(1982) 8175.
[181] Y. Mimura, N. Imamura, T. Kohayashi, Jpn. J. Appl. P l i w 17 (1978)
[I821 Y. Mimura. N. Imamura. T. Kobayashi. J. Appl. Phvs. 47 (1976) 368.
[183] S. Honda. K. Ueda, T. Kuzuda. J. Appl. Phys. 52 (1981) 2295.
[184] S. Tsunashimd, M. Tsuji, T. Kohayashi, S. Uchiyama. IEEE Trans. Mugn.
MAG-17 (1981) 2840.
[lX5] P. Hansen. M. Urner-Wille, 1 Appl. Phys. 50 (1979) 7471.
[186] M. Urner-Wille, P. Hansen. K. Witter, IEEE Truns. Mugn. MAG-16
(1980) 1188.
[187] M. Hartmann, P. Hansen, P. Willich, J. Appl. Phys. 56 (1984) 2870.
[188] N. Imamura, O w Butsuri 50 (1981) 1225.
[189] N. Imamura, S. Tanaka, E Tanaka, Y. Nagao, IEEE Truns. Mugn. MAG21 (1985) 1607.
[190] H. Kobayashi, T. Ono, A. Tsushima, Appl. Phys. Lett. 43 (1983) 389.
[191] T. Chen, M. Mansuripur, R. Malmhill, Proc. SPIE In/. Sor. Opt. Eng.
382 (1983) 260.
[192] T. Chen, R. Malmhall, G. B. Chdrlan, J. Appl. P h p . 53 (1982) 2356.
[193] T. Chen. D. Treves, Pror. S P l E Inr. Soc. Opt. Eng. 329 (1982) 264.
11941 R. Krishnan. M. Porte. M. Tessier. J. P. Vitton (Eds.), Polyni. Prepr.
( A m . Chem. Soc. Div. Polym. Chem.) 29-2 (1988) 232.
[195] T. Tokunaga, M. Harada, M. Ohkoschi, S. Honda, T. Kuzuda. I N T E R M A G Con/: 1986, FC-08.
[196] S. Honda, M. Ohkoschi, T. Kuzudd, IEEE Trans. M a p . MAG-22 (1986)
I1971 T. R. McGuire. IEEE Truns. Mugn. MAG-22 (1986) 1224.
[198] M. Hartmdnn. T. R. McGuire, Phys. Rev. Lett. 51 (1985) 1194.
[I991 S. Tanaka, Y. Nagao, F. Fdnaka, Kinzoku Hyomen G@ztsu38 (1987) 439.
12001 Y. Togami, IEEE Trans. Mugn. MAG-18 (1982) 1233.
[201] Y. Kudo, A. Muto. N. Hard, M. Birukdwa, H. Kawabata, Jpn. Null.
Tech. Rep. 1988, 28.
[202] J. J. Cuomo. R. J. Gambrno. J. Vac. Sci. Twhnol. 12 (1975) 79.
[203] a) J. F. Dillon. R. B. van Dover, M. Hong, E. M . Gyorgy. S. D. Alliston,
J. Appl. P h j x 61 (1987) 1103; b)Y. Nagao. S. Tanakd, F.Tanaka, N.
Imamura, Jpn. J. Appl. Phys. 21 (1982) L509.
[204] N. Imamurd. Proc. ‘’3 Int. Conf Ferriter Kyoto 1980, 798 (Center for
Academic Publications Japan, Tokyo 1981.
[205] D. Jeanniot. J. Desserre. J. Appl. Phys. 54 (1983) 2820.
[206] S. Tanaka. N. Imamura, J. M a p . Mugn. Muter. 35 (1983) 205.
[207] a) I. Sander, M. Urner-Wille. Proc. SPlE In!. Soc. Opt. Eng. 382 (1983)
240: h) K . Ohta. A. Takahoshi, T. Deguchi, T. Hyuga, S. Kohayashi. H.
Yamaoka, ibtd. 382 (1983) 252.
I2081 T. Deguchi, T. Hyugd, S. Kohayashi. A. Takahashi. K. Oh. H. Yamaoka,
Shupu Giho 22 (1982) 57.
[209] S. Uchiyama in Y. Sakurai (Ed.): Muprto-optic Properfreso/ Temury und
Quurriwwry Rurr Eurtli-Trunsition Metal Amorphous Films (Jpn. Annu.
Rev. Electron. Comput. Telecommun. 12 (1986)).
[210] a ) A . P. Malozemof. A. R. Williams, K. Terakura, V. L. Moruzzi. K.
Fukamxhi, J. Mugti. Mugn Muter. 35 (1983) 192: h) G. A. N. Connell,
;hid. 54-57(1986) 1561:c)G. A.N.Connel1.S. J . 0 h . J . Allen. R. Allen.
J. Non-Cr,wt. Solid\ 61 & 62 (1984) 1061
A n g ~ w Chem.
Int. Ed. Engl. 28 llYX9) 1445 1471
121I ] H . Tiujimoto, M. Shoiyi. A. Suito, S. Matsushita, Y. Sakudi, J. Jpn. Appt.
Mugn. Soc. 7 (1988) 119.
12121 D. Harzberg (Ed.): Optical Memory News Decemher 1988, Rothchild
[241] A. W. Smith. Appl. Opt 13 (1974) 795.
12421 S. R. Ovshinsky. US-Pat. 3 530441 (1970). Energy Conversion Devices
[213] M Tdkahashi, N. Okta, S. Tdkayama, Y. Sugita. M. Yoshihiro, K. Shigematsu. IEEE Trans. Mugn. M A G - 2 2 (1986) 931.
12141 S. Tdkayama. T. Niihara, K. Kaneko, Y. Sugita, M. Ojima, J. Appl. Phys.
I2431 a) S. Asai, E. Maruyama, Suppl. Jpn. J Appl. Pl~xs.40 (1971) 172; b) A.
Hamada, T. Kuroso. M . Saito, M. Kikuchi. Appl Phy.s. Let/. 20 (1972)
61 (1987) 2610.
12151 P. Hansen. J. Appl. Phys. 62 (1987) 216.
12161 a) B. G. Huth, IBM .
Res. Dev. 18 (1987) 100; b) M. Mansuripur,
C;. A N . Connell, J Appl. Phys. 55 (1984) 3049: c) Proc. SPIE l n l . Soc.
Opr. Eng. 420 (1983) 222; d) M. Mansuripur, M. F. Rudne, IEEE Trans.
M u p . M A G - 2 2 (1986) 33.
[217] M . Sato. K. Habu. T. Oyamd, IEEE Trans. Mugn. M A G - 2 3 (1987) 2614.
[ Z l S ] F Tanaka. Y. Nagdo. N . Imamurd, IEEE Truns. Magn. M A G - 2 0 (1984)
12191 M. Akihird, T. Sdtch. J. Tdda, T. Satoh, IEEE Trans. Mugn. M A G - 2 2
(1986) 92X.
12201 S . Takayama. F. Kirino. Y. Suzuki. S. Okamine, N. Ohta, IEEE Truns.
Mugn. M A G - 2 3 (1987) 2611
12211 M . Sato, S. Tdtsukawa, N. Tsukane, T. Mitani, M. Toba, Jpn. J. Appl.
P1iv.s. 26 (1987) 19.
12221 Y. Suzuki. S. Takayama, F. Kirino, N. Ohta. IEEE Trans. Mugn. M A G - 2 3
(1987) 2275.
[223] S. Chikazumi: Physics of Magnetism, Wiley. New York 1964.
[224] M. Mansuripur, M. F. Ruane, M . N. Hormstein, Proc. S P l E Int. Sue.
Opr. Eng. 52Y (1985) 25.
12251 a) Y. Nishihura, T. Katayama. Y. Yamaguchi, S. Ogawa. T. Tsushina, Jpn.
J. Appl. P h w . 17 (1978) 1083; b) R. B. van Dover, E. M . Gyorgy. R. P.
Frankenthal, M. Hang, D. J. Sicono1fi.J. Appl. Phvs.59 (1986) 1291;~)T.
Mlzoguchi. G. S. Cargill, ibid. 50 (1979) 3570.
[226] H. Hibst, Angew. Chem. 94 (1982) 263; Angew. Chem. lnt. Ed. Engl. 21
(1982) 270.
12271 A. Watada. H . Nakamura, H. Machida. IEEE Trans. Mugn., Mug. 23
(19x7) 2961.
12281 H. L. Glas. J. H. W. Liaw, J. Appl. Phgs. 49 (1978) 1578.
12291 M.Matsuoko, Y Hoshi, M. Ndoe, S. Yamanaka, IEEE Trans. M a p .
M A G I 8 (1982) 1119.
12301 R. Ahrenkiel, T. Coburn, IEEE Trans. M a p . M A G 1 1 (1975) 1103.
12311 J. W. D. Martens. A. B. Voermaus. IEEE Trans. Mugn. M A G - 2 0 (1984)
[232] H. Matsuda, H. Iwasdki, M . Kaneko, K. Aso, Jpn. Kokai Tokkyo Koho
J P 63-47908 (1988): Sony Corp.: Chem. Abstr. 109 (1988) 203521.
12331 J. W. D. Martens. W. L. Peters. P. Q. J. Nederpel, J. Appl. Phys. 55 (1984)
(2341 P. Hansen. J:P. Krumme, Thin Solid Films 114 (1984) 69.
12351 W. Tolkdorf. C:P. Klages, Thin Solid Films 114 (1984) 33.
12361 J.-P. Krumme, V. Doormann, P. Willich, J. Appl. Phys. 57 (1985) 3885.
12371 Y Yokoyama. S. Tsukahara, T. Tanaka, J Mugn. Mugn. Muter. 35 (1983)
[2381 Y. Yokoyamd. S. Tsukahara, T. Tanaka, IEEE Trans. Mugn. M A G - 2 0
(1984) 1039.
12391 a ) M. Abe. M. Gomi, Oyo Burstrri.57 (1988) 723; b) K. Inomata, Denshi
Zutrw 27 (1988) 40.
[240] B. R . Brown. Appl. Opt. 13 (1974) 761.
Angeil'. Cii(w. Inr. Ed. Engl. 28 (1989) 1445-1471
9; c) P. Chaudhari, S. Herd, D. Ast, M. H . Brodsky. R. J. von Gutfeld. J.
Non-Cryst. Solid.s 8.- 10 (1972) 900.
[244] J. Feinleib. J. de Neuville, S. C. Moss, S. R. Ovshinsky. Appl. Phys. Lerr.
18 (1971) 254.
12451 R. J. von Gutfeld, P. Chaudhari, J. Appl. Ph.y.s. 43 (1972) 4688.
12461 K. Weiser, R. J. Gambino. J. A. Reinhold, Appl. Phys. Lptr. 22 (1973) 48.
(2471 A. Saito, T. Maeda. H. Sugiyama, Y. Tsumoda. Appl. Opt. 27 (1988) 4274.
12481 K. Suzuki, DE-OS 3802679 (1988). Toshiba Corp.
12491 M. Ojima, T. Nakao. H. Sukeda, N. Okta, H. Yasuoka, T. Nishida, M.
Terao. Pror. S P l E lnt. Soc. Opr. Eng. 899 (1988) 154.
[250] M. Chen, K. A . Rubin. Proc. SPIE h i . Sot.. Opt. 6 i g . 1078 (1989)
[251] P. K . Raychaudhuri, Proc. S P l E lnr. Soc. Opr. Eng. 1078 (1989) 157.
[252] F.S. Jiang, L. S. Hau. F. X . Cyan. Proc. SPIE Int. Soc. Opt. Eng. 107H
(1989) 165.
12531 a) S. C. Stinson, Chem. Eng. N r w 66 (1988) No. 42, p. 30; b) W. J. Tomlinson, Appl. Opr. 23 (1984) 3990; c ) J. Hibino, E. Ando, Jpn. Kokai
Tokkyo Koho J P 63-273 688 (1988), Agency of Industrial Science and
Technology; Chrm. Abstr. 110 (1989) 222674: d) K. Ichimura. N. M o m ,
M. Sakurdgi, N. Aaki. Jpn. Kokai Tokkyo Koho J P 63-222 174 (1988),
Agency of Industrial Science and Technology: Chem. Ahstr. 110 (1989)
125 523.
[254] a) M. Eich, B. Reck, H . Ringsdorf, J. H . Wendorff. Proc. SPIE Int. Snc.
Opt. Eng. 682 (1987) 9 3 ; b) S. Tdzuke, Kokuguku 12 (1988) 85.
[255] Proc. I n / . Symp. Future Electronic Devices (Bioelectric and Molecular
Electric Devices), Tokyo 1985.
12561 W. Siol, U. Terbrdck, DE-OS 3708428 (1988). Rohm GmbH.
12571 W. Dabisch, P. Kuhn. S. Miiller, K. Narayanan, DE-OS 29 07 352 (1980).
[2%] T. Potrawa, H . Langhais, Chem. Bet.. 120 (1987) 1075.
[259] G. Holhngsworth, J. D. Barrie. B. Dunn, J. I. Zink, J. Am. Chcni. Sue. / I 0
(1988) 6569.
[260] a) W. J. Merz, J. R. Anderson. Bell Lab. Rec. 33 (1955) 335: b) M. Niitzuma, R. Sato, Ffrroelectrics 34 (1980) 37.
[261] K. Dransfeld, R. Pott, G. Khmpf, EP275881 (1988), Bayer AG.
12621 P. S. Murthy. K. A. Khngensmith, J. Michl, J. Appl. Polym. St.;. 31 (1986)
12631 P. Gutlich, S. Decurtins. H. Spiering, A. Mauser, DE-OS 36220005
(1988), Bayer AG.
[264] J. Friedrich, D. Haarer. Angew. Chem. 96 (1984) 96; Angew. Chem. lnt.
Ed. Engl. 23 (1984) 112.
12651 E. Kratzig, R. Orlowski, V. Doormann, M. Rosenkranz, Pro(. S P l E Inr.
Soc. Opt. Eng. 164 (1979) 33.
[266] Jpn. High-Technol. Monitor 7 (1989) 14.
12671 a) A. Renn, R. Locher. A. Meixner, M . P. Wild, J. Lumin. 38 (19117) 37;
b) Y. Jino, T. Tani, M. Sakuda, M. Nakahara. K. Fukuda, Chrm. Phys.
Lrrr. 140 (1987) 76: c) M. P. Wild. A. Renn, Pror. S P I E lnt. Soc. Opt.
Eng. 910 (1988) 61.
[268] H. Six1 in M. Borrisov (Ed.): Molecular Electronics, Singapore Press
1987, p. 11.
Без категории
Размер файла
6 498 Кб
data, optical, material, storage
Пожаловаться на содержимое документа