Headlines in Photonics Spectra over the past few years have included: “New Promise for Holographic Storage,” “Holographic Storage Takes a Step Toward Viability” and “Low-Power, Time-Domain Holography Holds Promise.” But we have yet to see an article about a holographic data storage product that has made it onto the market. Since the 1960s, holography has been hailed as the ultimate data storage technology. Researchers promised the ability to store up to a terabyte of data on a small storage medium with read/write speeds of up to 1 Gb/s. Back then, finding the right material was a challenge. Several years later, when many of the material problems were being solved, the next obstacle was finding cost-effective optical components. Today, components such as spatial light modulators and blue laser diodes are available, but the holographic data storage industry faces more hurdles, not the least of which is money. In the last year, three companies have disappeared, and several others are having difficulty finding funding. Will holographic data storage ever make it to market? The shrinking window Hans Coufal, manager of science and technology at IBM’s Almaden Research Center in San Jose, Calif., is not optimistic. He believes there’s a chance that holographic data storage will never happen. “The window of opportunity is getting smaller,” he said. “Other technologies are improving.” IBM is carrying out its own research into the technology, but also is keeping a watchful eye on developments being made by start-ups and research groups elsewhere. “We are evaluating all materials about the globe,” Coufal said. “It is a big step from a material to a piece of commercially viable recording media, and several companies have made tremendous progress.” The materials under investigation can be divided into three categories — crystals, polymers and glasses. They all change the local refractive index when illuminated with two intersecting laser beams. For a while, photorefractive crystals received much attention. One problem is that the light used to read holograms also erases them, so holograms must be fixed into the crystal, usually by heating. Furthermore, photorefractive crystals are generally not very sensitive and require high-power lasers. The German start-up Optostor was investigating holography using this method, but it failed to find funding and is selling off its assets. According to Coufal, IBM still uses the crystal lithium niobate for testing, but he doesn’t believe that it’s a commercially viable product for holographic data storage. The promise of polymers Two types of polymers can be used for holographic data storage: photopolymers, where bonds form or break on illumination, and photochromic polymers, where side-chain realignment occurs. The latter technology has the advantage of being reversible, making it ideal for rewritable storage. Swedish company Optilink investigated this technology but, like Optostor, failed to find sufficient funding and folded. Of the two polymer technologies, the photopolymers show the most promise. When irradiated with light of a certain wavelength, these materials break or form bonds in a reaction that is usually irreversible. Although this is a sensitive technique, stability in the dark over long periods and material shrinkage are critical issues. Organic photopolymers are often subject to aging processes caused by stresses that build up in the material during recording or by residual reactive species left behind after recording. Erasure may occur because of residual thermal diffusion of the molecules that record the hologram. Two US start-ups have made remarkable progress using photopolymers. InPhase Technologies in Longmont, Colo., is commercializing a two-chemistry system originally developed at Bell Labs, whereas Aprilis Inc. in Maynard, Mass., is commercializing technology originally developed at Polaroid Corp. InPhase’s technology is based on free-radical polymerization, a process in which bonds form when the storage medium is illuminated. Its medium is made from a mixture of two independently polymerizable, yet compatible, chemical systems — an approach that the company claims produces a highly sensitive medium in millimeter-thick optically flat formats. Recording discs are formed by in situ polymerization of one of the components to create a matrix, which remains fixed and does not shrink when holograms are written. The other component, which is photosensitive, remains unreacted and dissolved in the matrix. When this component absorbs a photon, free-radical polymerization forms bonds. This is an irreversible reaction, but the company is also working on a rewritable material. “We believe our first product will be on the market in 2005 and will be aimed at the enterprise and corporate sectors,” said Bill Wilson, chief scientist of InPhase. He described the product as having a read/write speed comparable to that of an optical disc. The speeds in the first-generation products won’t be great, he said, but will be sufficient, as the real advantage lies in the ability to randomly access the data. He believes that the first consumer product will be a ROM because it will be impossible to copy, not because of the larger storage capacities attainable. Aprilis based its technology on cationic polymerization in polymers containing cyclohexene oxide groups with siloxane spacers. Imaging initiates polymerization in light-struck regions. The company claims that this can occur with less shrinkage of the material, because during the reaction, rings that require additional free volume open, compensating for the shrinkage that usually accompanies the polymerization process. Glenn Horner, vice president of business development at Aprilis, foresees no problems regarding sensitization at blue wavelengths because, as he describes it, “Our media have been sensitized for us in the mid-green region of the optical spectrum.” The company’s current choice of lasers is based on the requirements for high-quality page-based recording. Horner indicated that only lasers that operate at 532 nm appear to deliver the combination of power, beam quality, size and cost that meets the imaging requirements. However, even with the aforementioned improvements, the main problem for polymer materials is shrinkage. If a hologram has been stored and the material has changed shape in the process, it becomes more difficult to find that hologram during reading. Glasses, on the other hand, do not have this problem. They also have a much smaller thermal expansion coefficient than polymers do, which is important in disk drives, where temperatures can soar. Glasses intrinsically have better optical properties than polymers, which naturally scatter light. The newest competitor in the race to market this storage technology is Cambridge, UK, start-up Polight Technologies Ltd., which plans to commercialize the use of inorganic glasses. Aprilis and InPhase have demonstrated storage densities of about 1.9 and 1.2 GB/cm2, respectively. Polight has not yet demonstrated the capacity of its materials, but it has made use of its chemistry background (as a spin-off of the chemistry department at Cambridge University) by developing a medium that it claims meets all the demands of holographic data storage. Unlike any other material in development, its storage media are chalcogenide glasses — amorphous materials that contain Group VI elements. The glass approach CEO Michael Ledzion sees the relatively low cost of red lasers as an advantage because Polight has a material that will work with both green and red light (Figure 1). The glasses, dubbed Holonide, store holograms because they are semiconductors. When the material absorbs a photon with energy equivalent to its bandgap, local structural deformations take place, inducing a large local change in refractive index. This change can be reversible, which means that Holonides can be used in rewritable storage solutions. Figure 1. Polight Technologies’ materials are sensitive to green and red light. Holograms also can be stored irreversibly in the material by using photons with higher energies, thus causing the bonds to break. And Holonides can be used to multiplex holograms (store many in the same space), further increasing the data storage capacity. The main advantage polymers have over chalcogenide glasses is that they are more sensitive. The absorption of one photon sets off a chain reaction so that the effect of one photon is multiplied many times. In chalcogenide glasses, one photon induces one change, and many more photons are needed to induce more change. However, this means the reaction is easier to control. “Our materials are as sensitive as the least-sensitive polymers,” Ledzion said. “But the fact that our reactions are more controllable is an advantage.” He believes that his company will have a product on the market by mid-2005 and that it could work with several drive manufacturers to develop products for varying markets. Drive development It’s common in this market to find one company that concentrates on developing the storage medium while a second develops the drive. The one exception is InPhase. “Our core technology is the storage medium, but we also have the intellectual property and expertise to build a drive,” Wilson said (Figure 2). Figure 2. InPhase is one of the few holographic data storage companies developing both the recording medium and the drive technology in-house. The development of holographic drives is concentrated in Asia, with most of the DVD companies either developing their own drives or evaluating other drives and materials. These companies are understandably secretive about their developments, and information is hard to come by. Optware Co. in Yokohama, Japan, however, is a start-up that is already selling its collinear holographic media analyzer system. The inventor of the minidisk, Hideyoshi Horimai, started the company. Senior vice president of sales and marketing, Fuji Tanaka, explained that holographic data storage systems are often bulky and complex but that Optware’s technology keeps the system compact. He attributes this to a unique servo system that also has reduced the size of the optical pickup and has removed vibration isolators. The analyzer is also upwardly compatible with CDs and DVDs. Figure 3. Optware’s collinear polarized holography takes this fundamental optical configuration. The company says its drive is more compact than other designs. Unlike other holographic techniques, it allows the use of optical disk servo technologies, because the reference and signal beams are bundled on the same axis (Figures 3 and 4). The company also has been evaluating Aprilis’ materials and achieving positive results (see Photonics Spectra, April 2003, p. 37). Figure 4. These reconstructed images using collinear polarized holography were produced without (top) and with (bottom) a polarizing beamsplitter. The signal-to-noise ratio is considerably poorer without polarization because of the effect of the reflected reference beam. Road map for storage It is clear that big advances have been made, but will holographic data storage be on the market by 2005, as so many of these companies promise? “If you look at holographic data storage in isolation, the advances over the last 10 years have been phenomenal,” said Barry Schechtman, executive director emeritus of the US-based Information Storage Industry Consortium. “However, mainstream storage technologies have also advanced rapidly. Hard-disk drives, for example, have smashed through what were thought to be fundamental limits. These technologies have the same density, capacity and data transfer rates as HDS [holographic data storage], but they are cheaper.” He believes that, when the technology does reach the marketplace, it will be an improvement over existing technologies by only a few orders of magnitude. “HDS will only succeed if it makes bigger advances than conventional technologies.” “Basically,” Schechtman said, “I think it is going to be tough. If HDS does not come onto the market by 2005, other technologies will overtake, even though HDS looks better on paper. If I were a personal investor, I would invest some money in HDS, but not my whole nest egg.” How Holographic Data Storage Works Today’s CDs, DVDs and hard-disk drives store information as bits on the surface of the storage medium. In holographic data storage, an entire page of information is stored at once as an optical interference pattern throughout the volume of a thick, photosensitive optical material. This is done by intersecting two coherent laser beams within the storage material. The first (the object beam) contains the information to be stored; the second (the reference beam) is designed to be simple to reproduce. The resulting optical interference pattern causes chemical and/or physical changes in the photosensitive medium: A replica of the pattern is stored as a change in the medium’s absorption, refractive index or thickness. When the stored interference grating is illuminated with one of the two waves used during recording, the grating diffracts some of this incident light in such a fashion that the other wave is reconstructed. Holographic data storage is faster than conventional data storage for two reasons: Laser beams can be moved rapidly without inertia, unlike the actuators in disk drives; and information can be read and written in pages, so that large amounts of data can be retrieved in a relatively short time.