1. Introduction

Long-term massive data backup prefers using optical storage technique due to its low cost per megabyte, long archive life and disk removability [1]. Conventional bit-oriented optical storage such as DVD and Blue-ray Disk (BD) records data (bits) on a thin surface layer of disk and has reached its physical limitation. Although BD can hold data up to 25 GB on a single layer by using a 405-nm laser, it increases the data storage capacity by only a few times over DVD. Thus, it is not considered as a significant increase in storage capacity in view of employing the same fundamental bit-per-pit storage technology. It is unlikely to further reduce the laser wavelength for next phase of improvement due to physics limitation and technical issues on both light source and optics design in ultraviolet or shorter spectral band.

Holography is a volumetric technique which has long been considered as a next generation optical storage technology because of its high density and fast transfer rate [2–4]. Traditional holographic multiplexing involves spatial displacement of recording beams, thus requires a completely new disk drive architecture that is not back compatible to the current bit-oriented optical disk format. This will cause countless data that have already been recorded by bit-oriented optical disks to be unreadable. To address this issue, micro-grating multiplexing [5–8] has been studied to take the advantages of both holography and bit-oriented storage by recording a reflective grating with a laser beam of multiple wavelengths at each spot. However, such multiplexing is limited by few laser lines available in the spectral sensitivity range of the recording media.

We recently proposed a method for recording microholograms with white light source [9], however the multiplexing was only simulated by sequentially overlapping individual gratings recorded by single wavelength due to the lack of effective wavelength-multiplexing methods. In addition, the sizes of the micro-grating in previous studies are still too large comparing to the pit size on the existing optical disks, and it has not been demonstrated if the micro-grating multiplexing technique is compatible with the conventional disk techniques for readout and recording.

In this study, we constructed an electrically programmable wavelength combiner and demonstrated that the micro-grating multiplexing can be obtained by using a single beam containing multiple wavelengths coded by the wavelength combiner. More importantly, we have successfully recorded a micro-grating array with individual diameter close to the diffraction limit while the transverse spot size of the grating is well maintained through a 15-micron thick recording medium by using a focal-depth-extended hybrid diffractive and refractive lens. In addition to significantly increasing storage capacity, the importance of this study is that we demonstrated the feasibility of recording microholograms with multiple bits per pit that are back compatible with the conventional optical disk storage technology.

2. Fabrication of recording/readout lens with extended focal depth

Previous study using conventional lens [5–8] for holographic multiplexing has problem on keeping spot size constant throughout the medium volume due to beam divergence. In fact, it is impossible to record a single micron-scale hologram throughout depth exceeding the Rayleigh range with a conventional lens which obeys the following well-known relationships:

where Δx is the minimum resolvable spot size in the transverse dimension, Δz is the depth of focus (DOF) of a focused beam by the lens, f _# is the f number of the lens, and λ is the wavelength. k ₁ and k ₂ are constants that depend on the incident beam truncation and the criteria adopted [10].

The two relationships imply that there is a conflict between reducing spot size and increasing DOF. For example, to achieve the minimum resolvable spot size of 0.6 µm yielded by the Rayleigh criterion at 500 nm wavelength with a f _#=1 lens, the DOF is less than 1 µm. Increasing focal depth Δz simultaneously enlarges minimum resolvable feature size Δx. This problem can be overcome by using our novel hybrid achromat which is the integration of a diffractive optical element and a refractive optics specially designed to produce a focused beam with an elongated DOF. The operating principle has been described in detail [11].

In this study, we optimized our extended-DOF hybrid lens for microholographic multiplexing applications to achieve a diffraction-limited spot size while extending the focal depth to cover the entire 15 µm film thickness that we used. The specially-designed diffractive optical element in the extended-DOF lens generates a long range of pseudo non-diffractive ray which combines effectively with a refractive optics lens to diminish any chromatic aberrations in the desired spectral band. Utilizing a hybrid refractive-diffractive device configuration simultaneously preserves the favorable properties of both the diffractive element (long focal depth) and the refractive lens (low chromatic aberration and high energy concentration). The fabricated hybrid lens works in the entire 400 nm to 700 nm waveband. We measured the DOF of our new lens by imaging the focal spot at various axial distances. The results shown in Fig. 1 indicate that the DOF is larger than 30 µm with a transversal resolution less than 1 µm. It has a significantly extended DOF while preserving high lateral resolution as compared to a conventional refractive lens with the same f _#.

 

Fig. 1. Images at different focal planes of the hybrid f _#=1 lens: (a) 2.990 mm, (b) 3.000 mm, (c) 3.010 mm, and (d) 3.020 mm from the lens. The measured DOF is larger than 30 µm.

Download Full Size | PDF

3. Construction of wavelength combiner

To successfully record wavelength-coded holographic data, the recording beam should be able to be coded with selected wavelengths and desired intensities for each individual wavelength. Lasers are common for recording holograms due to their excellent coherence with narrow spectral linewidth. However, it is impractical to find many cost-effective compact lasers with evenly distributed laser wavelengths in the visible spectral band. White light sources are inexpensive and have plenty of wavelengths for spectral multiplexing. If we can electrically modulate all the spectral lines in a spectral band of material response, the wavelength-coded recording beam will be created using a low-cost white light source.

We have designed and constructed a compact wavelength combiner to support our coded high-density holographic storage application. The configuration of the wavelength combiner is based on an asymmetrical crossed Czerny-Turner spectrometer structure [12]. The constructed combiner is shown in Fig. 2. Incoherent white light from a Xenon lamp (or Tungsten) is used for our study. A bifurcated fiber bundle is used for both the input of white light and the output of the spectral selected light with discrete wavelengths. Inside the wavelength combiner, the white light from the optical fiber is collimated to a reflection-type grating by a concave mirror. The diffraction beam is collected and focused by using another concave mirror with the same focal length onto a reflection spatial light modulator (SLM). The computer generated binary bar codes representing to be stored data sequences are displayed on the SLM for controlling the reflection of the spectral components. We align the system so that the reflected light by the SLM with discrete wavelength propagates in opposite direction back to the optical fiber. The optical alignment ensures the terminal core of the input fiber to be exactly imaged back on the core of output fiber in the same bifurcated fiber bundle. With the SLM, the intensity of each wavelength component can also be adjusted by controlling the reflectance of the SLM pixels. By using this wavelength combiner, we can code the white light in real-time with desired output wavelengths for the recording of microholograms.

 

Fig. 2. Configuration schematic of the constructed wavelength combiner.

Download Full Size | PDF

 

Fig. 3. Spectrum of a wavelength-multiplexed recording beam obtained from a white light source by using the constructed wavelength combiner.

Download Full Size | PDF

Figure 3 shows an example of light source with discrete wavelengths generated by using our programmable wavelength combiner. The wavelength-coded light transmits through the optical fiber to the optical head of the hybrid extended-DOF lens for holographic recording. Our wavelength combiner has spectral resolution of about 1 nm in the visible region (400–700 nm). A Holoeye LC-2002 SLM was used in the wavelength combiner for feasibility demonstration. It has a frame rate of 60 Hz. This frame rate is fairly low for practical applications that require faster data transfer rate. The transfer rate can be improved by using a SLM with faster switching rate, such as digital micro-mirror array. The fast development of superluminescent LED and new high-power white light source will also facilitate the transfer rate improvement of our technology. In fact, the broad superluminescent LEDs with high optical output power (up to 30 mW in single-mode fiber) are already commercially available. This may greatly enhance the application of our technology.

4. Construction of compact recording/readout head

To increase storage density and demonstrate the compatibility of micro-holographic storage to conventional optical disk storage, it is important to study the feasibility of recording microhologram with minimum diameter throughout the recording medium thickness by using our hybrid lens. In this study we used a DuPont holographic film (HRF-800X001-15) as recording medium due to its high diffractive efficiency and relatively flat spectral response in the visible spectral region. The DuPont photopolymer film consists of polymeric binders, acrylic monomers, and plasticizers, along with initiating systems including initiators, chain transfer agents, and photosensitizing dyes. The dyes absorb light to interact with the initiators to begin photopolymerization of monomers, resulting in refractive index modulation. The proper choice of components allows tailoring of the film physical properties and the magnitude of index modulation for specific applications.

For this study, the films are formulated to work best in the recording of reflection holograms. The holographic photopolymer is 15 µm sandwiched between a Mylar polyester substrate and a removeable cover sheet. Before exposure, we removed the cover sheet and attach it with the photopolymer side directly onto a clear mirror. In order to obtain ultra-fine focal spot, a single-mode fiber is used as output fiber of the wavelength combiner. The light emerging at the output fiber end is collimated to the hybrid lens which focuses the beam vertically onto the recording film.

 

Fig. 4. Schematic of the recording/readout architecture of our design. M’s: mirrors, G: transmission grating, BS: beam splitter, L: collimating lens, RD: recording disk, EDFL: extended depth of focus lens, BF: bifurcated fiber bundle, CL: cylindrical lens, AD: autofocus detector

Download Full Size | PDF

Figure 4 shows a schematic of the recording/readout architecture of our design. The experiment setup is simple and compact; however, the optical alignment is difficult since we have to precisely focus the lens with the middle of the extended focal depth on the mirror surface so that the incident and reflective beams within the focused region are well overlapped throughout the film thickness. To ensure proper beam focusing for recording, we have also designed an auto-focus system similar to that used in the DVD recording head. Instead of using an autofocus magnetic actuator, we used a piezo actuator system to achieve focus adjustment with better precision and larger range.

By modifying a commercial CD readout head, we implemented a compact recording/readout head of the design. A picture is shown in Fig. 5. With our design, only one beam is required for grating recording and there is minimal requirement on light coherence since the recording optical path length is short, about double of the film thickness.

 

Fig. 5. Implementation of the designed recording/readout architecture by modifying a CD readout head.

Download Full Size | PDF

5. Microhologram recording

The recording exposure is typically 5 to 100 mJ/cm² for the DuPont films with wavelengths ranging from visible to near-IR [13,14]. To obtain microholograms with diffraction-limited spot size, we need to control the exposure dosage so that the recording film is only sensitive to the energy at the focal center rather than in the beam side-lobes. For our experiments, using exposure of 10–20 mJ/cm² yields the best results. After exposing to the recording beam, the film is cured for about 5 minutes by using a UV light (UV75 Light, Thorlabs Inc.) to stabilize the recorded microholograms. The initial diffraction efficiency (reflectance) of the microholograms is about 10%. By heating the gratings at a temperature of 120 °C for 1 hour, the diffraction efficiency reaches nearly 95%.

 

Fig. 6. Experimental result of reflective-type micro-hologram array with submicron spot diameter using the Dupont photopolymer.

Download Full Size | PDF

By using our recording system, we have successfully recorded a reflective-type microhologram array as shown in Fig. 6. The diameter (FWHM) of each individual microhologram is in submicron level close to the diffraction limit. These microholograms are recorded well through the thickness of the recording medium with the maintained submicron size. This can be observed under a high-resolution microscope by changing the focusing depth while the recorded microholograms are illuminated by a white light source. We also found that the reflected spectral curves from the microholograms have the same spectral FWHM selectivity bandwidth as that with a larger diameter (such as 1 mm) recorded in the whole volume of the same DuPont films. Since the spectral selectivity bandwidth of the grating depends on the recording thickness (i.e., grating interaction length), the same spectral resolution further demonstrates that the microholograms have been fabricated through the whole thickness of the recording film.

Figure 7 shows the spectral readout result of a multiplexed microhologram recorded by using the wavelength-coded light from the wavelength combiner whose spectrum is shown in Fig. 3. We can see that the readout result matches the optical code of the recording beam except that there is a several nanometers blue-shift in the readout spectral peak corresponding to each recording wavelength due to the recording film shrinkage during the photo-polymerization process. The spectral FWHM selectivity bandwidth is about 5 nm by using the 15-µm DuPont film. Since the spectral selectivity bandwidth Δλ of Bragg reflection grating depends on the film thickness L, i.e.Δλ=λ ²/(2n _eff L), the number of wavelengths to be multiplexed in the same storage pit can be further increased by using a thicker recording media. In addition, our technique supports recording of multiple stacked layers. Thus, it is possible to record micro-holograms individually throughout a much thick recording film for further increase of storage capacity.

 

Fig. 7. Spectral read-out of a micro-hologram recorded by using the wavelength-multiplexed light beam shown in Fig. 3.

Download Full Size | PDF

5. Conclusions

To explore the practicality of microhologram data storage technique, several significant improvements have been performed in this study. We designed and fabricated a hybrid diffractive and refractive achromat with extended depth of focus and successfully recorded a micro-hologram array with diffraction-limited individual spot size maintained throughout the thickness of recording medium. An electrically programmable wavelength combiner was constructed in which a white light source was adopted. By modifying on a commercial CD readout head, we configured a compact micro-hologram recording/readout system that is compatible to existing disk storage technology. Base on the wavelength combiner and the recording/readout system, wavelength-multiplexed micro-holograms were recorded and recovered. The presented results demonstrate the practicality of our novel storage architecture. The technique has potential to become the next generation optical disk storage due to its high storage density, parallel and fast data access, and back compatibility with existing single-beam recording/readout storage architecture.