We present an optical write / read system for high density optical data storage in 3-D. The microholographic approach relies on submicron-sized reflection gratings that encode the digital data. As in conventional optical data storage, the physical limitations are imposed by both the diffraction of light and resolution of the recording material. We demonstrate resolution-limited volume recording in photopolymer materials sensitive in the green and violet spectral range. The volume occupied by a micrograting scales down by the transition in the write / read wavelength. Readout yields a micrograting width of 306 nm at 532 nm and 197 nm at 405 nm. To our knowledge these are the smallest volume holograms ever recorded. The recordings demonstrate the potential of the technique for volumetric optical structuring, data storage and encryption.
© 2011 OSA
Advances in data storage are driven by the increasing density of stored information. Optical information storage is based on a light-matter interaction: The digital data is encoded as an optically detectable change in a recording material. In conventional optical storage, this change is a simple quarterwave groove in the reflective layer . The optical resolution limit, known as the “Abbe barrier”, has dictated progress and development among three generations of optical disk technology. The approach to higher data densities was straightforward – reducing the size of data marks by shortening the wavelength and by increasing the numerical aperture. In its third generation, based on a 405 nm laser, optical disk storage definitively encounters the physical limits imposed by diffraction of light. Meanwhile, new techniques that use more than the surface of a flat disk are under investigation. Multilayer [2–6], holographic [7–9], microholographic [10–14] and multidimensional [15–17] approaches are expected to exploit the entire volume of a storage medium. In addition to the spatial dimensions, optical multiplexing techniques allow superimposing multiple states that differ in wavelength, angle of incidence, phase or polarization.
In contrast to binary page-wise holographic data storage, the microholographic approach capitalizes on its fundamental compatibility with the established optical disk technology [11,12]. The data is stored holographically in three dimensions but bit-wise in tracks and layers similar to those of a standard optical disk. The reflectivity of a photosensitive material, typically a holographic recording photopolymer, is locally varied by recording submicron-sized reflection gratings. Cross-talk, typical in page-wise recording, is eliminated by the bit-wise nature of the recording and readout process: At any point in time the focused light beam illuminates only one microholographic bit feature. Multiplexing techniques open additional paths to storage densities beyond the resolution limit imposed on 2D optical data storage. Multilayer recording is a simple method of spatial multiplexing that relies on the depth localization of holographic microgratings. Starting from data densities comparable to DVD or BluRay the microholographic multilayer approach targets the TeraByte capacity range. At the same time, it allows a cost-effective and downward compatible technology implementation as the drive system has most optical and optoelectronic components in common with a standard optical drive.
A major focus in the practical realization is the physical data structure and resulting data density in a single microholographic layer that can be recorded and read independently from another layer. These parameters represent a basis for the volumetric data density achievable by multilayer recording. In our microholographic system bit features are volume reflection gratings created by two tightly focused, counterpropagating laser beams. They interference pattern is strongly modulated in its 3D envelope and wavefronts. Transferring such a light pattern into a nanometer-scaled volume index alteration of an organic material system is one of the challenging issues in microholographic data storage. The readout mechanism is based on the interaction of a focused laser beam with the stored index modulation.
In this paper we address the submicron-volume localization and investigate the potential of the microholographic approach in terms of physical data density and capacity. We briefly discuss the implications and limitations inherent to the recording material and present the optical write / read system optimized for high-density recording. We take a critical step forward in demonstrating microholographic storage in the resolution-limited regime in two photopolymer systems sensitized to green and violet light, respectively.
2. Recoding / readout configuration
The design of our optical system for high-density microholographic storage is based on a single-beam path, using a retroreflecting configuration. For recording a laser beam of adequate coherence is focused at a certain storage location. After passing through the recording medium and another objective it becomes retroreflected so as to generate a “second” write beam (Fig. 1 ). The grating formation takes place in the joint focal region of the two counterpropagating beams when a photosensitive polymer is exposed to their interference. The intensity pattern is transferred into a medium by the process of polymerization and subsequent diffusion of monomers [18,19]. The majority of holographic polymers are linear in their photoinitiating system response . Consequently, the index modulation in illuminated regions is expected to exactly mirror the interference pattern. Ideally, a micrograting induced in a photopolymer is spatially constrained to the focal volume of the write beam. The optical theory predicts the lowest limit for the transversal and longitudinal dimensions of microgratings, which is defined by the wavelength of the laser light and by the numerical aperture of the optical system. In fact, it is not the recorded feature itself, but rather the optical signal generated by diffraction at this feature that dictates the storage performance. The readout signal generated by the interaction of light with the recorded feature yields the reconstructed information and is the figure of merit of an optical storage system.
In our microholographic system, readout is performed by a laser beam identical to the original write beam without its retroreflected part. When the read beam exactly overlaps with the recorded grating fringes, the resulting Bragg diffraction yields a reflected beam propagating in the opposite direction. Spatially resolved detection reveals the spatial distribution of the recorded index modulation of a micrograting. A major development effort concerns design of a write / read system that ensures both resolution-limited 3D-localized recording and highly sensitive readout. In particular, the system has to provide an exact overlap of the two counterpropagating beams in the recording process. Displacements in the incident and reflected beam path result in a focal dislocation that effects an enlarged, distorted grating structure as a consequence. In the readout step, the same degree of precision is required when the wavefronts of the read beam interact with the recorded localized grating fringes. Deviations from the full overlap lead to Bragg mismatch and concomitant losses in diffraction efficiency, i.e. readout signal. The spatial selectivity of the readout system is improved by confocal filtering of the diffracted light.
In addition to the write / read setup, the recording material is a crucial element of the entire storage system. The interaction of the optical system and polymeric storage medium governs the storage performance in terms of data rates, density and capacity. In particular, for resolution-limited localization of microgratings the optical index modulation of a photopolymer has to accurately reproduce the light interference pattern that is of high spatial frequency and strongly modulated in 3D. Microholographic storage requires highly homogeneous, highly sensitive photopolymers that record and exhibit a stable volumetrically localized refractive index change coincident with the recording light interference pattern and allow multiplexing to increase data density. The optical quality of the organic material system in terms of optical resolution and scatter is decisive for the signal quality and bit error rates . The chemical formulation and recording chemistry of a photopolymer dictates its dynamic response to pulsed illumination and can enable fast, dynamic and submicron-localized recording of 3D grating structures . Among photosensitive polymerizable materials different chemical mechanisms can be involved in the recording process. Even though the basic steps of polymerization and resultant mass transport by diffusion are common for most holographic photopolymers, there are differences in their chemistry of recording, optical characteristics, material dynamics and stability [20,22,23]. Photopolymer materials used in this work are cationic ring opening polymerization (CROP) systems developed by Aprilis [24–26].
3. Optical write / read system
Our optical write / read system for microlocalized recording is shown in Fig. 2 . In this work two different laser systems are used: A 100 mW frequency-doubled Nd:YAG laser for recording and readout in green-sensitized photopolymers and an external-cavity diode laser at 405 nm for violet-sensitized materials. The laser emission is modulated by an acousto-optic modulator. Typical exposure times range between 100 ns and 100 µs. Focusing optics comprises two identical microscope objectives with a high numerical aperture of 0.75 that focus the counterpropagating beams into the photopolymer. The objectives are corrected for spherical aberration. The Gaussian beam is truncated by the objective aperture. The resulting flat top beam yields diffraction-limited spot size at the given wavelength.
In the incident beam path, the recording beam passes a spatial filter, becomes focused by the first objective at a certain storage location within a photopolymer sample and is then collimated by the second objective and propagated to a lens. The second write beam is generated by reflection of the incident beam at a retroreflector unit, which consists of a lens focusing onto a mirror. This configuration provides for the reflected beam overlapping with the incident write beam on the same optical axis. The specification of the objective lens requires an angular alignment smaller than 70 µrad. Using retroreflection only the longitudinal position of the second microscope objective has to be corrected during the system operation. The setup is switched from recording to readout mode by blocking the reflected beam path with a mechanical shutter.
During readout, light reflected at the recorded microgratings passes the same spatial filtering section as the read beam and is then directed by a polarizing beam splitter into the detection path. The spatial filter is designed as a confocal filtering unit, which significantly improves the sensitivity and resolution of the optical readout system. After passing the confocal filter, the reflected beam is coupled onto a photodetector. An important advantage of this configuration is that the confocal filter is always optimally aligned as a part of the recording / readout beam path.
The combined performance of these two components, retroreflector and confocal filter, is a key feature of the optical system design. Together they ensure the beam path alignment and also improve the spatial selectivity of detection in the write / read system. Note the retroreflector configuration implies losses in the index contrast as the reflected beam covers a larger distance within the recording material than the incident write beam. Until the two focal spots overlap at a bit location, the second beam passes twice the distance within the photopolymer sample, passes twice through the multi-element second objective and is being reflected by the retroreflector unit. Typically, the total power loss, including both the material losses and system-related losses, is approximately 30% as compared to the power of the incident beam part. This causes loss of contrast or modulation depth in the writing interference pattern and thereby losses in achievable diffraction efficiency for given exposure fluence. An alternative optical configuration with two symmetrical, but separate, beam paths can avoid this kind of loss. In this case additional components are necessary to align the two beam paths and to keep this alignment stable at all times during the system operation. In spite of the losses, the retroreflector configuration is favorable because of its coincident beam path and the advantage in adjusting and maintaining the overlap of the two diffraction-limited beams.
The noise level in the tested Aprilis CROP photopolymer materials is on the order of 10−5 diffraction efficiency units, which makes the diffraction efficiency range between 10−4 and 10−3 (where SNR is high enough for electronic devices) particularly relevant for resolution-limited recording. To determine the noise level, a polymer sample is scanned by a focused laser beam. The light scattered back in the axial direction of the probe beam is captured at a photodetector.
For the opto-mechanical system part, a major concern is high-precision control of the photopolymer position. Photopolymer media samples are provided by Aprilis as two inch coupon media that consist of a 300 µm thick polymer layer sandwiched between two antireflection-coated glass substrates. The polymer sample is mounted on a three-axis translation stage that is driven by a servo motor and is optically controlled. The spatial positioning resolution is 50 nm with a maximum translation speed of 50 mm/s. Microgratings are recorded either in the dynamic or quasi-dynamic regime (“stop-and-go-mode”). Readout is always performed dynamically by scanning the sample with a probe beam that is identical to the incident write beam. The readout signal is generated by reflection of the read beam at recorded microgratings.
4. Resolution-limited recording
Figure 3 summarizes the recording / readout results on single microgratings written in a green-sensitized Aprilis photopolymer and in a violet-sensitized sample at 405 nm. The micrograting shown in Fig. 3(a) is recorded at 532 nm and NA of 0.6. The spatial distribution of the reflected light is dictated by the Bragg matching of grating fringes and wavefronts of the focused read beam. The spectral transition from 532 nm to 405 nm in write / read wavelength is a logical step in the system development. At the same time the numerical aperture of the focusing objectives is increased from 0.6 to 0.75. The transversal size of a micrograting created in the Aprilis photopolymer follows the expected λ / NA relationship. Figure 3(b) shows the diffraction efficiency of a micrograting recorded at 405 nm and NA of 0.75. The comparison in Fig. 3(c) highlights the resolution-limited downscaling of lateral dimensions: the FWHM diameter of a ‘violet’ micrograting shrinks to 197 nm as compared to 306 nm at 532 nm write / read wavelength.
At this degree of spatial localization, the recording material plays a decisive role. Recording in the submicron range requires an adequate recording chemistry and chemical composition of the photopolymer. A micrograting written at 405 nm with a nominal refractive index of 1.5 has a fringe spacing of 135 nm, so the maximum depth scale of the localized polymerized volume of the polymer chains has to be shorter. Otherwise the local refractive index variation would not be limited to the exposed regions; the contrast would be washed-out and holograms become larger than the light spot. The interaction between the incident light and the photopolymer system is affected by a number of parameters that might be roughly classified into three blocks: optical system, recording material, and exposure conditions. The amount and duration of the exposure can be crucial factors for the volumetric localization of the polymerization reactions and the resultant localized chemical segregation related to the diffusion-driven mass transport of monomers and binders. This requires a material-specific optimization of exposure conditions as the photopolymer response is shown to be strongly dependent on the exposure fluence.
5. Recording exposure optimization
In practice, several optimization steps are necessary to find a suitable combination of exposure parameters for a given photopolymer sample and optical specification. A typical exposure series is shown in Fig. 4 . Such investigations are performed without confocal filtering in the readout section so that the entire index-modulated volume can be observed. The exposure response shows a threshold below which the material response is too weak to be separated from the background noise. Linear recording is achieved through the plotted regime of the dynamic range, which is above the threshold and below saturation. In the exhibited linear recording range, increasing exposure fluence leads to higher diffraction efficiency but the volume modulated by photopolymerization becomes larger. Over-modulation effects occur when the deposited energy is too high. In this case the recorded structure becomes much larger than the write beam spot with a distinctive doughnut shape as predicted by coupled-wave modeling . The material response varies among different photopolymer systems and parameters may change even for different samples of the same material . Microlocalization is shown to be ideally achieved in the so-called low-modulation range. In this range the exposure fluence is high enough to ensure error-free readout, but it does not yield maximum values of the diffraction efficiency. Diffusion of monomers, polymerization and subsequent modulation of the refractive index take place only in a confined central part of the focal volume where the light intensity is highest. Low-modulation microgratings are therefore significantly smaller than the high-reflective ones. They can become strongly localized in volume and arranged densely in tracks and layers.
We determine optimized exposure fluence as a compromise solution for the requirements of strong volume localization and sufficiently high diffraction efficiency. The latter is critical for microholographic recording. With only a few grating fringes effectively contributing to diffraction from the formed microstructure, the theoretically achievable efficiency at the available maximum index change Δn of a recording material is limited. Using a coupled wave model for Bragg diffraction at a micrograting , a maximum diffraction efficiency of 10−3 can be achieved for an index change on the order of 10−3.
In practice, a recording setup adds optical losses as well as material losses that diminish the maximum achievable efficiency. Diffraction efficiency and overall losses are summarized by the signal-to-noise ratio (SNR). SNR directly affects the bit-error rate (BER) and is an important figure of merit of a storage system. Usually, the main source of noise in a microholographic storage system is the recording material itself. Particularly for submicron-sized structures, the most limiting factors are scattering noise as well as local variations in diffraction efficiency. The lowest acceptable diffraction efficiency of a single grating is limited by the SNR required by the optoelectronic detection system. The trade-off between the SNR and micrograting size becomes obvious from the example exposure series in Fig. 4. Submicron-sized gratings exhibiting diffraction efficiency of 10−3 are written at 532 nm and NA 0.6 with a total fluence about 1 J/cm2. At this recording condition, the reflected light can be properly separated from noise while the micrograting is confined to 500 nm in its lateral direction and 5 µm in its depth. With higher exposure fluence, SNR improves but the grating structure becomes larger so the SNR improvement comes at the expense of decreasing data density.
6. Resolution-limited in-track recording
Next, we investigate the spacing between microgratings written in tracks. The influence of the recording material becomes even more critical when many microgratings are recorded close to each other. Data density achievable in a single layer is governed by both the minimum structure size and the required spacing of data marks. In Fig. 5 single tracks of microgratings are written as bit-pattern in ‘stop-and-go’ mode: In the recording process the photopolymer sample is translated by the positioning system until storage location is reached, and then stopped for the duration of exposure. Readout is performed dynamically by continuously scanning the exposed area. The spacing between adjacent microgratings is reduced from 500 nm (left) to 350 nm (right). Even at spot distances shorter than the wavelength, individual microgratings are clearly distinguishable but contrast becomes weaker and a monotonic decline in diffraction efficiency occurs with subsequent recordings along the track. At larger spacing, 500 nm or more, the variation in DE is smaller and is attributed to be mainly an effect of nanoscaled inhomogeneities in the material composition. At small grating spacing, the adjacent exposed regions begin to overlap and therefore cause exposure and polymerization in the neighboring interaction volumes. The gratings spaced by 350 nm already share material dynamics since the polymerization process is no longer localized to the individual storage locations. In our observation, densely written multiple gratings influence each other and interact through the dynamic grating formation process.
Optical disk technology across three generations has brought optical data storage to perfection at the physical limits of resolution and diffraction. The experimental data presented herein shows that resolution-limited optical storage is also possible in 3D with submicron-sized volume gratings representing individual bit features. The storage performance is governed by the interaction between the focused laser beam and photopolymer material. The key issues in the design and development of the optical system are 3D microlocalization of recording, selectivity and sensitivity of readout. Much research fails to recognize the importance and implications of the storage medium for system design.
Though primarily developed for page-oriented holographic storage, Aprilis CROP photopolymers are shown to be capable of bit-wise volume localized recording. The recorded microgratings feature a lateral width of 306 nm in green-sensitive photopolymer media. The micrograting width scales down with the write / read wavelength to 197 nm at 405 nm. In both photopolymer media the recorded microgratings feature a width smaller than the write / read wavelength and are the smallest volume holograms ever recorded.
Microgratings with good SNR characteristics, recorded at exposure fluence ranging from 1 J/cm2 to 10 J/cm2, are arranged densely along a track. Error-free readout is demonstrated from microgratings with in-track spacing of 350 nm. The results achieved in the Aprilis CROP photopolymer media verify the potential of microholographic storage to overcome the fundamental limitations of the optical disk technology.
We acknowledge Dr. David A. Waldman, Aprilis co-founder and CTO, now founder of HolFocus, for developing and supplying the photopolymer media for microholographic recording, and the European Commission for financial support.
References and links
1. M. Mansuripur, The Physical Principles of Magneto-optical Recording (Cambridge Univ. Press, 1995), Chap. 1.
3. S. Hunter, F. Kiamilev, S. C. Esener, D. A. Parthenopoulos, and P. M. Rentzepis, “Potentials of two-photon based 3-D optical memories for high performance computing,” Appl. Opt. 29(14), 2058–2066 (1990). [CrossRef]
4. J. W. Perry, B. H. Cumpston, S. P. Ananthavel, S. Barlow, D. L. Dyer, J. E. Ehrlich, L. L. Erskine, A. A. Heikal, S. M. Kuebler, I.-Y. S. Lee, D. McCord-Maughon, J. Qin, H. Röckel, M. Rumi, X.-L. Wu, and S. R. Marder, “Two-photon polymerization initiators for three-dimensional optical data storage and microfabrication,” Nature 398(6722), 51–54 (1999). [CrossRef]
8. S. Homan and A. E. Willner, “High-capacity optical storage using multiple wavelengths, multiple layers and volume holograms,” Electron. Lett. 31(8), 621–623 (1995). [CrossRef]
9. H. J. Coufal, D. Psaltis, and G. T. Sincerbox, eds., Holographic Data Storage (Springer-Verlag, 2000).
10. H. J. Eichler, P. Kuemmel, S. Orlic, and A. Wappelt, “High density disk storage by multiplexed microholograms,” IEEE J. Sel. Top. Quantum Electron. 4(5), 840–848 (1998). [CrossRef]
11. S. Orlic, S. Ulm, and H. J. Eichler, “3D bit-oriented optical storage in photopolymers,” J. Opt. A, Pure Appl. Opt. 3(1), 72–81 (2001). [CrossRef]
12. R. R. McLeod, A. J. Daiber, M. E. McDonald, T. L. Robertson, T. Slagle, S. L. Sochava, and L. Hesselink, “Microholographic multilayer optical disk data storage,” Appl. Opt. 44(16), 3197–3207 (2005). [CrossRef]
13. M. Dubois, X. Shi, C. Erben, B. Lawrence, E. Boden, and K. Longley, “Microholograms recorded in a thermoplastic medium for three-dimensional data storage,” Jpn. J. Appl. Phys. 45(2B), 1239–1245 (2006). [CrossRef]
14. K. Saito and S. Kobayashi, “Analysis of micro-reflector 3D optical disc recording,” Proc. SPIE 6282, 628213 (2007). [CrossRef]
15. D. Day, M. Gu, and A. Smallridge, “Rewritable 3D bit optical data storage in a PMMA-based photorefractive polymer,” Adv. Mater. (Deerfield Beach Fla.) 13(12-13), 1005–1007 (2001). [CrossRef]
16. X. P. Li, J. W. M. Chon, S. H. Wu, R. A. Evans, and M. Gu, “Rewritable polarization-encoded multilayer data storage in 2,5-dimethyl-4-(p-nitrophenylazo)anisole doped polymer,” Opt. Lett. 32(3), 277–279 (2007). [CrossRef]
19. G. Odian, Principles of Polymerisation, 4th ed. (John Wiles & Sons Inc., 2004).
20. R. T. Ingwall and D. A. Waldman, in Holographic Data Storage, H. J. Coufal, D. Psaltis, G. T. Sincerbox, eds. (Springer-Verlag, 2000), Chap. Photopolymer systems.
21. S. Orlic, E. Dietz, T. Feid, S. Frohmann, and C. Mueller, “Optical investigation of photopolymer systems for microholographic storage,” J. Opt. A, Pure Appl. Opt. 11(2), 024014 (2009). [CrossRef]
22. L. Dhar, M. G. Schnoes, H. E. Katz, A. Hale, and M. L. Schilling, in Holographic Data Storage, H. J. Coufal, D. Psaltis, G. T. Sincerbox, eds. (Springer-Verlag, 2000), Chap. Photopolymers for digital holographic storage.
23. B. Kippelen, in Holographic Data Storage, H. J. Coufal, D. Psaltis, G. T. Sincerbox, eds. (Springer-Verlag, 2000), Chap. Overview of photorefractive polymers for holographic data storage.
24. D. A. Waldman, R. T. Ingwall, P. K. Dhal, M. G. Horner, E. S. Kolb, H.-Y. S. Li, R. A. Minns, and H. G. Schild, “Cationic ring-opening photopolymerization methods for volume hologram recording,” Proc. SPIE 2689, 127–141 (1996). [CrossRef]
25. D. A. Waldman, C. J. Butler, and D. H. Raguin, “CROP holographic storage media for optical data storage at greater than 100 bits/μm2,” Proc. SPIE 5216, 10–25 (2003). [CrossRef]
26. D. A. Waldman, E. S. Kolb, C. Wang, “DHD™ CROP holographic storage media for advanced optical data storage,” Optical Data Storage (ODS), OSA Technical Digest Series WDPD 4–7 (2007).
27. H. Kogelnik, “Coupled wave theory for thick hologram gratings,” Bell Syst. Tech. J. 48, 2909–2947 (1969).