Click for next page ( 39


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 38
4 Opportunities in Storage and Display INTRODUCTION Optical storage and display subsystems are used for input and output of data in information processing systems. Optical printers, which use optical beams to write on drums that transfer the image to the paper, are also part of the input/output subsystem but are not treated in this report. There is already a large U.S. industrial and military market for these optical subsystems, but they also have a strategic importance for future information- handling systems. As information-processing systems grow in complexity to handle an ever-broader range of applications as well as more sophisticated applications, the demand to store massive amounts of data and to display and print large amounts of data becomes critical. The optical technologies used in these subsystems promise the capability to store and display more information than their mechanical, electronic, and magnetic counterparts. Therefore, having a strong technology base in these optical and optoelectronic materials and de- vices is potentially critical to having a leading-edge information-processing system. The word data is to be interpreted in this section in its broadest sense, that is, information relating to entertainment, culture, education, and business as well as the more common scientific connotation of the word. Thus these information systems affect virtually all human activity. 38

OCR for page 38
OPPORTUNITIES IN STORAGE AND DISPLAY OPIICAL STORAGE Importance of Optical Storage 39 Measured in the hundreds of billions of dollars, information storage is an industry that isvitalto this country's military, cultural, and economic well-being. One type of storage, archival storage, is dominated by printing on paper (95 percent of the words written are on paper). The other type of storage is on- line, rapid-access storage, which is currently dominated by magnetic storage media such as tapes and disks. On-line storage itself is more than a $50 billion per year industry and is growing at a rate of about 15 percent per year. The long-term trend, as cost permits, is to store all information electronically on- line because of the inherent ease in accessing, manipulating, and disseminating the information. This chapter will thus emphasize on-line storage applications. Optical storage media store information more densely than their magnetic counterparts and thus potentially at lower cost. Therefore optical storage tech- nology could displace large segments of the magnetic storage market. Optical storage also has the potential of storing information much more densely than paper and thus could also displace this market segment, although in the near term optical storage technology is too expensive for many archival applications. In computer data storage, for example, an optical disk can store 500 times more data than a floppy disk of comparable size. For high-speed data access, an optical disk can store 50 times more data than the competing magnetic rigid disk. With "jukebox" configurations where the disks are interchanged on a disk reader, the storage can be greatly expanded at the expense of access time. A more detailed description of this application has been given by Chi (1981~.i In the area of archival storage, one optical disk is capable of storing the entire Encyclopaedia Britannica of 43 million words and 24 thousand pictures. Ten optical storage boxes could store all of the information currently contained in the Library of Congress. This type of system has been described in detail by Bartolini (1982~.2 In the entertainment/arts field, the compact disc--an optical disk--is already replacing the phonograph record. In this case, the larger storage capacity coupled with digital encoding has been utilized to improve audio fidelity. The exceptionally large storage capabilities of optical media make possible a third, unique application based on combining digital and video information on the same disk. As an example, consider a future "interactive encyclopedia" where a stored subject consists of words, pictures, and video movies to fully describe the subject, complete with user-activated cross-references or elabora- tion capability. This application today is a $100 million per year industry that is projected to be a $1 billion industry by the 1990s. Besides having an economic impact, this technology could lead to new concepts in learning, literature, or the arts, that go beyond what is available ~ a written textbook. For example, the

OCR for page 38
40 PHO TONICS "interactive novel," which allows its reader to select from many potential plots and thus to take part in the story, is made more feasible with a large, quickly branchable, storage medium such as optical storage. Since the storage of data is so critically important in todays world and since future generations of data storage are likely to include optical storage, the United States must gain and maintain a leadership position in optical storage technology. This critical need is amplified even further when one considers that optical storage can displace many types of storage technology in various fields, from computers to literature to audio/video. As with many of the optical technologies, the dominant development activity appears to be in Japan. Storage Systems Storage systems can be divided into three types. The fast is the multiple read and write system, currently a $70 billion per year industry that is dominated by magnetic media. Optical storage systems of this type are under develo`pment at many major computer and communications companies worldwide. The second type is archival storage where media are written once by the user but can be read many times. Todays market is dominated by magnetic tapes, micro- f0, and printed paper. The audio and video disk technology has already been reengineered in Japan and introduced into this storage market. The third type of system is the read-only system, where replicas of a master disk are distributed to users. In this case, only preexisting data are available to the user, who cannot generate data on this system. This market is immense and difficult to quantify as it includes photographic film (including movies), phonograph records, and printed paper. As mentioned above, optical storage has penetrated the audio application segment of this market and is starting to penetrate the printed paper segment. For the read/write applications, high storage density and data access speed comparable to those of the magnetic hard disk are key technical require- ments, whereas in archival storage applications, the cost and permanence of the stored information are critical. Technology Elements An optical storage system consists of a number of key technology elements. These are the storage media, the reading and writing head, and the electronic interfaces and system design and software.S,6 As shown schematically in Figure 4.1, semiconductor lasers are used as a source of light that is focused onto a very small spot on the optical storage media. The small spot causes a material change when the information is to be written. The spot is read by a change in the light reflected off the small spot (of the order of 1 micron in size). The spot

OCR for page 38
OPPORTUNITIES IN STORAGE AND DISPLAY / - - - - - - Disk rotation - .~ - - - 3, Laser beam 41 .aser movement FIGURE 4.1 Diagram of optical storage system. The same laser may be used for both reading and writing. sequence is converted to valid data or information sequences by the storage system protocols and sent to the data processing system or to a specialized "librarian" computer. Media There are a number of different materials for recording these dots of infor- mation. Table 4.1 lists some of these materials, the storage processes, and the relative advantages of each. The main approaches used today are ablative for archival storage and phase change for read/write writing, coupled with magneto-optical readout. Key issues in the media today are the read and write optical power required, media lifetime, and media cost. Current methods for writing information on optical disks use the laser as a heat source for removing small spots of material (the ablative process) or for causing a phase change in a small spot of the material. The photonic nature of the laser beam is utilized only in that the light wavelength sets the limits on the size of the written spot. Because of the heating requirement, high-power lasers (of the order of 50 mW)

OCR for page 38
42 TABLE 4.1 Various Optical Storage Media and Characteristics PHO TONICS Type Storage Advantages Disadvantagcs Read-only Injection molding Pit formation Stability Low replication cost High mastering cost Write-once Ablative Pit formation Sensitivity Optical write power (Te alloy) Most developed Potential for bit error Data permanence Ablative Bubble formation High SNR Stability (dye polymer) Low manufacturing cost Potential for bit error Phase change Optical reflectivity Lower write power Lifetime change in contact Low SNR overcoat Reversible Magneto-optic Magnetic domain Reversibility Low SNR switching Most advanced technique Mcdia cxpcnsc Kerr effect readback Mcdia passivation Phase change Optical reflectivity High SNR (potential) Lin~itcd reversibility change Simpler optical system Higher write power Single-pass overwrite Dye polymer Materials flow Cost Limited reversibility Stability Lifetime Read/writc speed are critical for the writing process. The reading of optically stored information is based on a change in reflectivity or a change in polarization of the reflected light; this process requires less laser power, only enough to obtain an adequately noise-free signal. Read/Write Head The key element of the read/write head is the semiconductor Ejection laser. Critical issues are the optical output power, device lifetime at high power, shorter light wavelength for smaller spots and high recording density, modula- tion speed, and optical beam quality. Expensive, bulky gas lasers are available for writing but are limited primarily to the equipment that makes "masters." Adequate injection lasers for small spot writing (about 50 mW of single-mode optical power with adequate product life) are only begunning to be developed.

OCR for page 38
OPPORTUNITIES IN STORAGE AND DISPLAY 43 Read-only injection lasers are available, with Japanese industry the primary supplier. The reflected light is read by a photodetector, usually segmented into four quadrants, for use in tracking a spot's position and improving the signal strength with respect to system noise. The beam is focused by a series of optical lenses and beam splitters and a miniature beam-positioning system based on "voice coil" actuators. The key attributes are efficiency in delivering power to a small spot and low mass so that the head can be rapidly repositioned. Storage System Architecture The architecture of an optical storage system refers to the way the data are formatted onto the disk, the way the data are read from the disk, and the way the computer system interacts with the optical storage system. For example, in read-only memory, contiguous data sequences are rearranged into noncon- tiguous dot patterns on the disk and coded with error-correcting codes in order to reduce the number of erroneously read bits by more than 7 orders of magnitude. These formats are often critical to the practicality of a technology for an application. For example, a factor retarding the use of an ideal write- once storage with infinite storage capability and zero cost is the lack of good ways to store various copies of a particular piece of information so that the proper (i.e., the most recent) copy is retrieved efficiently. (It appears that the human brain is ahead of computer science in this respect.) The panel will not cover these architectural aspects of optical storage systems in detail, except to point out where they seem to be controlling the development of the hardware. Competitive Environment Initial research in optical storage systems was carried on by a number of U.S. and European labs: Philips on laser tracking concepts and error correcting coding, INCA on high-powered lasers and archival systems, IBM on magneto- optic read/write media, and IBM/MCI and AT&T Bell Labs on ablative media. Products were developed but never introduced in the commercial marketplace (e.g., Storage Technology Corporation's gigabit file system), largely because U.S. industry could not identify near-term large markets. Today the primary research and development and product work is done in Japan at the large electronic companies. Current products are compact audio and video disks and archival systems for personal computer and workstation applications. Read/write systems are near the product introduction stage. In this as in other areas reviewed in this report, the lack of leadership in optoelectronic and photonic manufacturing technologies in the United States represents a serious threat to a significant, emerging area of economic competition.

OCR for page 38
44 PHO TONICS Key Enabling Technology Needs In a number of areas in optical storage, technological improvements are critical to product competitiveness. Technology elements that are in a position to be incorporated into products in the next 5 years are called enabling technol- ogy. Elements that need a longer, sustained research effort are considered in the next subsection. In the area of read/write heads, there is clearly a need for improved high-power laser sources. Maintaining good beam characteristics and demonstrating long life at high power are key issues. High power may come from more efficient single devices or from arrays of interacting lasers. There is good reason to expect that these enabling improvements are possible. New advances in the quality of compound semiconductor growth, such as the use of molecular-beam epitaxy (MBE) or metal-organic chemical vapor deposition (MOCVD), are allowing newlaser structures to be explored. Lasers with quan- tum well confinement, new alloy compositions, and nonabsorbing mirrors are being researched; the result could be lower laser threshold currents, improved efficiency, shorter wavelength, and the ability to operate a device at higher power without damage. Planar-processed laser structures might lead to better ways of coupling the optical beam from the laser or improving the beam properties. An alternative approach to high optical power is to pump small Nd-YAG rod lasers with diode lasers. This approach is not as attractive from a wavelength, packaging density/mass, or efficiency viewpoint. Manufacturable packaging technology for lasers and the optical head is an important component of assuring a good optical source. Migration from the assembly of miniature but discrete optical elements to planar-processed, self-aligned elements is critical to offering a competitive technology. Low-mass read/write heads are important to fast data access. The in- tegration of signal source and detection and perhaps some of the processing functions into an OEIC chip could provide a low-mass head with enhanced capability and functions for optical storage systems. Advances in laser sources as well as in optoelectronics require a high degree of sophistication and materials control in the processing of compound semiconductors. These areas will require sustained research and development activity, even though some forms of integration can be developed today. Another aspect of faster data access lies in multiple-track reading and multiple "platter" systems. The integration mentioned above helps packaging density in these types of systems but needs to be extended to cover arrays of function, for example, arrays of read/write lasers that are simultaneously and independently accessing tracks on the disk.

OCR for page 38
OPPORTUNITIES IN STORAGE AND DISPLAY 45 The key problem in read/write applications is the availability of a reversible, high-contrast material. Improvement of existing materials must be vigorously pursued. The photo-induced change in optical reflective properties must occur rapidly and be stable for long periods of time without degradation. Finally, it must be possible to repeatably read and write the memory without degradation. The challenge here is similar to that for the optical logic applications: a large optical change triggered by a small optical impulse (a large optical nonlinear effect) is desired to clearly delineate two optical states. The problem is compounded by the need for large areas of this nonlinear material (compared to a logic chip), but the regions do not have to be interac- tive or unbred. Currently available systems are based on funs that can be cycled between stable and metastable phases with accompanying reflectivity or Magnetic Kerr effect changes. The quality of such funs is limiting current applications, and significant advances in the storage media would have a major impact on read/write technology. Key Research Areas A longer-term need is the advance of laser technology, particularly small semiconductor injection diode lasers. Here a key development is the migration of lasers to shorter wavelength, either by new laser materials or by frequency doubling of existing lasers. Basic materials research is needed in nonlinear optical materials for frequency doubling, with emphasis on com- patibility with low-mass read-write heads. In the area of storage media, basic materials research is fundamental to future development. In this context, a variety of other phenomena have been explored for optical information storage. These include photochemical hole burning, holographic storage, and ferro-electric polarization-change storage. Although attractive in some respects, photochemical hole burning requires ultralow temperatures for long-term stability. Holographic storage mechanisms maybe of considerable importance in image processing (discussed in Chapter 3) but may not be competitive with the already high information storage density capability of optical compact disks. Ferroelectric ceramic materials may become important if the recent rapid progress continues in developing thin-f~ materials with good storage characteristics and low voltage requirements. However, these materials also appear to be of greater importance in image processing due to the inherently fast read/write speed of this medium. While all these areas are long-shot technologies, research into new methods of optical storage should be encouraged. All the current methods, as illustrated in Table 4.1, utilize thermal heating by a laser beam for writing information, and

OCR for page 38
46 PHO TONICS this inherently requires appreciable power. Other approaches based on photonic effects in materials should be possible, for example, an absorbed photon causing either a photo-induced charge-transfer reaction or a photo- induced change in molecular conformation leading to a metastable change in optical properties. One might hope that other phenomena based on photonic effects in materials would lead to significant improvements in read/write head requirements or media stability. The ability to store exceedingly high densities of information in archival systems is coming and may lead to conceptually new applications. Advances may depend critically on nonhardware factors such as storage system design (software and architecture) and the storage-computer system interface. New architectures are needed to organize and access vast quantities of information at a high rate of speed. As software and application development has histori- cally occurred at a slower rate than development of hardware, it is important to encourage early work in this area by sponsoring seed research and by supplying early prototypes to system designers. OPTICAL DISPLAYS Importance of Optical Displays Displays take electronic information and convert it to images and text for human viewing. Since displays are viewed, they necessarilyinvolve optics. How- ever, their operation is better described by how they are addressed or written; in this sense they can be either electronic or photonic. The commercial market in displays was about $7 billion in 1986 and is projected to be over $9 billion by 1990. The market is dominated by the cathode ray tube (CRT) display, which is currently at about 75 percent of the market share. There is a growing market in flat panel display technology, driven by the desire for compact, low-voltage (light-weight, low-power-consumption) displays. Today's displays are addressed primarily by electrons and hence are electronic displays. Examples are the beam-addressed CRT and the matrix- addressed flat panel displays (Figure 4.2~. Photonic displays are addressed by light beams. An example is a display where a laser beam writes on a liquid crystal cell (thermal writing process). Although U.S. and European corpora- tions are involved in research and development and commercialization of displays, the main effort appears to be from the major Japanese electronic corporations. One reason for their dominance may be that the Japanese internal market is also the largest in the world.

OCR for page 38
OPPORTUNITIES IN STORAGE AND DISPLAY Addressing Hard Wired ~ I r r I Beam Production Type of of Light Beam Nonemissive | Emissive Photon | Electron 1 Liquid Crystal Light-Emitting Diode Laser-Liquid Crystal Cathode-Ray Tube Electrochromic Electroluminescent Electrophoretic Plasma Panel FIGURE 4.2 Examples of beam-addressed and matrix-addressed displays. Display Status Flat Panel Displays 47 The major activity in flat panel displays today is in electronically matrix- addressed panels. A voltage is applied to each picture element either to induce light to be emitted (emissive) or to cause a change in optical transmission or reflection of light from an external light source (non-emissive). As these displays are not "photonic" in the above defined sense, their characteristics and status will be covered only briefly. The relative advantages and disadvantages of some representative display technologies are summarized in Table 4.2. A more detailed description of these displays was given in a special issue of IEEE Spectrum (1985~. The liquid crystal display, matrix addressed, is the dominant flat panel technology (at about a 15 percent share of the total display market in 1987), primarily in applications that cannot use CRTs. Earlier problems of slow switching speed and low con- trast have been solved by active matrix elements involving thin-film transistors or diodes in conjunction with the liquid crystal at each picture element. High (30:1) contrast is now possible at video refresh rates and large viewing angles. The challenge remaining is to manufacture such displays at costs competitive with those of CRTs in order to penetrate the higher-volume market. In addition, new materials such as ferroelectric liquid crystals, which can respond

OCR for page 38
48 TABLE 4.2 Relative Advantages and Disadvantages of Representative Display Technologies Type Advantages Disadvantages Beam-addressed - Cathode ray tube High resolution High voltage Good addressability Large depth High contrast High ambient lifetime Flexibility Corner edge focus Color capability circuitry High luminous efficiency High maintenance cost Heavy Emissive flat panel Light-emitting diode Extremely fast Short persistence High resolution Poor luminous efficiency Rugged Brightness uniformity Reliable High peak currents Low voltage No blue light Expensive in large arrays Electroluminescent display Rugged Moderate luminous High contrast efficiency Inherent memory Moderate luminance Expensive drivers Non-emissive flat panel Liquid crystal display Electrochromatic display Passive display crystal Memory possible Very high resolution No contrast loss in ambient light Passive display High contrast Inherent memory Low switching voltage External illumination required Temperature range Addressing, multiplexing Viewing angle Contrast limitations External illumination required Difficult to matrix- address Electrode and electrolyte stability Slow switching speed PHO TONICS more rapidly to external fields and can exhibit memory, could have a significant impact on a variety of applications for this display technology. Plasma panel displays and vacuum fluorescent displays, utilizing a gas dis- charge to emit light from a picture element to the viewer, represent a slowly growing market (from 8 percent of the market in 1987 to 9 percent in 1990), primarily in specialized, high-information-content displays. Electroluminescent (EL) panels are also present in the marketplace.9~0 Materials research directed toward improved EL efficiency, better reliability, and full color would benefit this technology. A common problem in the "gas" displays is the high cost of driver electronics.

OCR for page 38
OPPORTUNITIES IN STORAGE AND DISPLAY Photonic Displays 49 Laser beams have been used to write on liquid crystals or photoconductive media to produce images, much as an electron beam writes on a phosphor in a CRT. Large-area and high-resolution displays have been demonstrated in specialized applications. Interactive Displays The interaction between the display and its viewer is distinct from how the display is made or addressed but may be impacted by photonics. An example is the eye-controlled display, where the orientation of the viewer's eyeball deter- mines the display's cursor position. Although the concepts and necessary technology are available, a successful product has not emerged. An exciting application would be a variable resolution display, where the portion of the display being focused on by the user is high-resolution (with greater information content), while the peripheral region has low resolution (minimum information density displayed). Key Enabling Technology Needs The key technology developments needed in displays are in the flat panel electronic displays. The panel does not currently see any key enabling technology elements in photonic displays that would allow them to displace existing electronic display technologies or open new fields of application. Key Research Areas There should be continued research in new materials and subsystem con- f~gurations, which might lead to a photonic display that is superior to its electronic counterpart in the long term. Potential advantages of using photonics could result from the fact that light beams do not interact, leading to novel mul- ti-beam addressing schemes. In addition, new photoexcited display media may have higher resolution or image-storage capability. The promise of the hologram as the key element of a three-dimensional display remains a tantalizing goal. New concepts are needed to advance photonic displays if they are to become competitive. Some of the technologies developed for optical

OCR for page 38
50 PHO TONICS information processing or optical storage may be applicable to photonic displays. REFERENCES Chi, C. S. 1981. High density-for disk memories. IEEE Spectrum 18:39. Bartolini, R. A. 1982. Optical recording: High density information storage and retrieval. Proceedings of the IEEE 70:589-597. Bruno, R. 1987. Making compact disks interactive. IEEE Spectrum 24(November):40-45. 4. Freese, R. P. 1987. Erasable optical disks. IEEE Spectrum 23 (Feb- ruary3:41-45. Bartolini, R. A., A. E. Bell, R. E. Flory, M. Lurie, and F. W. Spong. 1978. Optical disk systems emerge. IEEE Spectrum 15(August):20-28. White, R. 1980. Disk-storage technology. Scientific American 243 (August):138-148. 7. 1987. Electronic Display World (7~:7-8. Stanford, California:Stanford Resources, Inc. 8. IEEE. 1985. Special Issue on Display Technologies. IEEE Spectrum 22:52-67. 9. Tannas, Jr., L. E. 1986. Electroluminescence catches the public eye. IEEE Spectrum 23~0ctober):37-42. 10. Manuel, T. 1987. The picture brightens in flat panel technology. Electronics 60~11~:55-58. 11. Levine, J. L. 1984. Performance of an eye-tracker for office use. Computers in Biology and Medicine 14~1~:77.