1. technology and computing

© 2005 OSA/ISOM, ODS 2005
MB1
Holographic storage without holography: Optical data
storage by localized alteration of a format hologram
R. R. McLeod
Department of Electrical and Computer Engineering, University of Colorado, Boulder Colorado, 80309-425
Phone: 303-735-0997, Fax: 303-492-2758, Email: mcleod@colorado.edu
A.J. Daiber, M.E. McDonald, S. L. Sochava
Intel Communications Group, Intel Corporation, 8674 Thornton Ave, TB-2, Newark, CA 94560-3330
T. Honda
Canon, Inc., 23-10 Kiyohara-Koghodanchi, Utsunomiya-shi, Tochigi 321-3298 Japan
T.L. Robertson
University of California, Berkeley, Department of Physics, Berkeley, CA 94720-7300
T. Slagle
Foveon, Inc., 2820 San Tomas Expressway, Santa Clara, CA 95051
L. Hesselink
Department of Electrical Engineering, CIS-X Room 325, Stanford University, Stanford, California 94305
Abstract: We propose and demonstrate multi-layer storage in holographic photopolymer by locally
altering the reflectivity of a factory-written reflection hologram at the focus of a single objective lens.
Linear, two-photon and thermal writing mechanisms are demonstrated.
© 2005 Optical Society of America
OCIS codes:: (210.4590) Optical disks, (210.2860) Holographic and volume memories
1. Introduction
The evolution of optical data storage from the CD to the next generation of blue laser DVD products has steadily
increased capacity by reducing wavelength (from 780 nm to 405 nm), decreasing track pitch (Rayleigh resolvability
from 2.9 to 2.09) and increasing numerical aperture (from 0.45 to 0.85). Since transmission of most materials is
limited to wavelengths greater than roughly 350 nm and mechanical tolerances become intractable beyond 0.85 NA,
it appears that this trend may be near its limit. Storing at multiple depths is the only remaining method to increase
capacity. However, two significant problems strongly limit the number of layers. First, fabrication of a many-layer
optical disk with the required tolerances and cost is a significant challenge. Second and more fundamental, standard
write-once, read-many (WORM) storage materials change physical phase by heating the material with absorbed
laser power. Given a fundamental minimum layer thickness of ~4 nm and the requirement of sufficient absorption
to generate the minimum temperature change of ~100 oC, signal-to-noise ratio during reading of the lowest layer
limits the number of layers to roughly four [1]. Optical data storage with many layers thus requires an optical
storage material that can be fabricated as a homogeneous volume and which does not require the large absorption of
thermally-mediated writing. A candidate material is the class of photopolymers developed for holographic data
storage (HDS) that can be cast into one millimeter thick films, have low total absorption during writing, bleach to
nearly no absorption before reading and in which data are written via polymerization and subsequent component
diffusion to create a permanent index change [2,3].
2. System architecture
A mm-thick film of such polymer could be the basis for a multi-layer optical disk with the addition of a confocal
pinhole to reject reflections from nearby layers as shown in Figure 1(a). Unfortunately, index perturbations written
at the focus of the objective contain only low spatial frequencies and thus no signal is reflected back to the confocal
filter. However, if the polymer volume is first holographically patterned with a uniform reflection hologram, the
converging spherical wave exiting the objective will efficiently reflect (Bragg match) in an annulus whose radius is
determined by the grating pitch. Any perturbation to the grating reflectivity at the focus of the objective will be
detectable as a decrease in the normally high reflection, as shown in Figure 1(b). A 2D scan from this system is
shown in Figure 1(c) [4].
We demonstrate in the next section that the polymerization initiated at the focus of the objective by the same
linear absorption used to write the format hologram will locally erase the format grating. More importantly,
nonlinear writing methods such as two-photon or thermally-mediated changes that destroy or modify the format
grating are also demonstrated.
This architecture thus enables a traditional high-power write cycle followed by
many low-power read cycles but within many layers of an easily-fabricated volume. Two or three reflection
© 2005 OSA/ISOM, ODS 2005
MB1
holograms written with slightly different pitch or direction can be interfered to form a Moiré pattern which defines
tracks in radius and layers in depth in a single optical exposure. This is the analog of stamping servo tracks in a
single-layer R/W disk. Servo optics can then read these patterns and be used to drive radial and depth actuators to
lock the focus to any desired track within the 3D volume of the material.
Radial access
QWP
(a)
PBS
R/W
head
(c)
Laser
Depth
access
Disk
1 µm
Confocal pinhole
Data and servo detection
(b)
1
0
0
1
0
1
Fig. 1. Concept of the multilayer drive. Part (a) shows the optical layout which is a standard optical drive with the
addition of a confocal pinhole in the data and servo detection arm. The inset shows the calculated index profile in radius
and depth of a typical bit written via linear secondary polymerization. Part (b) shows the expected reflected intensity
which is normally high except where writing has locally degraded the hologram reflectivity. Part (c) shows a typical result
from the 5th layer of a six layer disk written via two-photon absorption. Bits are placed on a 2 x 2 x 20 µm grid.
We demonstrate in the next section that the polymerization initiated at the focus of the objective by the same
linear absorption used to write the format hologram will locally erase the format grating. More importantly,
nonlinear writing methods such as two-photon or thermally-mediated changes that destroy or modify the format
grating are also demonstrated.
This architecture thus enables a traditional high-power write cycle followed by
many low-power read cycles but within many layers of an easily-fabricated volume. Two or three reflection
holograms written with slightly different pitch or direction can be interfered to form a Moiré pattern which defines
tracks in radius and layers in depth in a single optical exposure. This is the analog of stamping servo tracks in a
single-layer R/W disk. Servo optics can then read these patterns and be used to drive radial and depth actuators to
lock the focus to any desired track within the 3D volume of the material.
3. Read/write results
Unlike traditional page-based or micro-hologram storage in which the optical exposure writes a hologram with
precisely defined pitch and shape, our goal now is to destroy or modify the existing hologram. This is a simpler task
and thus there are a number of mechanisms possible. Here we show read/write traces from four such methods that
fall into two general categories. No attempt to optimize the writing transfer rate has been made in these proof-ofprinciple demonstrations. All experiments use Polaroid (now Aprilis) ULSH-5B sensitized at 532 nm [2].
The first and simplest method, shown in Figure 2(a) is to erase the format hologram by linearly initiating
polymerization in the formerly unexposed dark fringes. In this experiment the format grating was written at 532 nm
but angle tuned for reflectivity at 670 nm. Forty tracks separated by 3 µm were written in five layers of data
separated by 20 µm. Due to the low sensitivity at 670 nm, each bit required 40 mW for 2 ms with a 0.5 NA focus.
The sample was then cured in white light before reading at 670 nm. Note that non-uniformity in the format
hologram results in a variation in the reflectivity of the “0” level.
Figure 2(b) shows data written into a similar format grating utilizing the unoptimized two-photon response of
the photo-acid generator (PAG) at 670 nm. The photoinitiator concentration has been reduced so that after
bleaching significant concentrations of PAG and monomer remain, removing any one-photon sensitivity. To
demonstrate this, the polymer sample was left in room lights for 33 days before writing. Due to the low two-photon
cross-section, each bit required 50 mW for 3 s but could be nondestructively read at lower power, as shown.
Figure 2(c,d) show two different versions of thermally-driven inelastic deformation of the format grating
causing it to tune towards increased (c) or reduced reflectivity (d). The format grating in this case is Bragg matched
© 2005 OSA/ISOM, ODS 2005
MB1
to 532 nm, fixed via room lights for one day and then UV exposed with the 200 J/cm2 of the 313 line of a mercury
lamp. Writing was accomplished via a 0.5 NA objective and 40 ns, one µJ pulses at 532 nm from a Q-switched
Nd:YAG laser. The data is from the central of 43 tracks written 2 µm apart and the lowest of five layers written 12
µm apart. The lower trace in figure 2(c) was read 65 hours later demonstrating that the distortion is not transient.
3.E-04
0.03
(b)
4.E-04
0.04
1 0 1 1 0 1 1 1
2.E-04
0.02
0 1 0 1
Reflectivity
Reflectivity, %
6.E-04
0.06
(a)
1.E-04
0.01
0.E+00
0.00
2.5
Reflectivity, %
2.E-04
0.02
0.E+00
0.00
20
2
0 0 1 1 1 0 0 1 0 1 1 0
25
30
35
40
0
45
5
10
15
20
25
0.8
0 hours
(c)
65 hours
(d)
0.7
0.6
0 1 1 1 0 0 1 0 1 1 0
0.5
1.5
0.4
1
0
0
1
1
1
0
0
0.3
1
0.2
0.5
0.1
0
0
0
5
10
15
X [µm]
20
25
0
5
10
15
X [µm]
20
25
Fig. 2. Read/write results for four different writing mechanisms. The mechanisms are: (a) Linear secondary polymerization, (b) twophoton secondary polymerization, (c) tuning towards Bragg matching with thermal deformation, and (d) tuning away from Bragg
matching with thermal deformation. See text for details.
4. Summary
We have proposed and demonstrated a new form of multi-layer storage in which linear or nonlinear writing
mechanisms locally modify the reflectivity of a factory-written reflection grating. Confocal filtering of the reflected
signal restricts sensitivity to the focus, enabling layer spacing as low as 12 µm. The ability to holographically
format the entire disk and the demonstrated nonlinear writing mechanisms offer the potential for very high capacity
holographic storage using only a single objective in a traditional optical disk layout.
Acknowledgments: This work was carried out while all authors were at Siros Technologies.
5. References
1
2
3
4
K. A. Rubin, H. J. Rosen, W. W. Tang, W. Imaino, T. C. Strand, “Multilevel Volumetric Optical Disk
Storage,” SPIE 2338 Optical Data Storage, 1994, p. 247-250
D. A. Waldman, R. T. Ingwall, P.K. Dhal, M.G. Horner, E. S. Kolb, H.-Y. S. Li, R. A. Minns, H.G. Schild,
“Cationic ring-opening photopolymerimization methods for volume hologram recording,” Proc. SPIE 2689,
pp. 127-141, 1996
L. Dhar, A. Hale, H. E. Katz, M. L. Schilling, M. G. Schnoes, F. C. Schilling, “Recording media that exhibit
high dynamic range for digital holographic data storage“, Optics Letters, V24, N7, 487-489, 1999.
L. Hesselink, “Three dimensional recording (3DR) technology,” Optical Data Storage, 2000 Conference
Digest, pp. 149-151