Open Access
Add to BibSonomy
Abstract
The multiplex technique increases the capacity of optical data storage, but the current reading throughputs is limited by the single-bit reading. We propose a fast readout method of multidimensional optical data storage using interference-aided reflectance spectral measurement to readout multiple bits information simultaneously. The multidimensional data is recorded in the photoresist layer on the disc with dielectric multilayer substrate by laser direct writing. With the designed interference layer inside the disc, the relation of thickness of recording layer and the peak shift of the reflected spectra has been built up. With different writing depths representing different bit of data, 2 bits and 3 bits unit information have been recorded and successfully read out at one exposure. This fast readout method is not only suitable for optical data storage by engineering the optical path length for example Blu-ray disc but also for super resolution optical data storage.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Optical data storage is one of the essential tools for the next generation of big data storage for its multidimensional storage, ultrahigh throughputs, ultrahigh density, ultrahigh security, ultralong lifetime and ultrahigh energy efficiency [1]. Multidimensional optical storage can increase storage capacity without being limited by wavelength and numerical aperture (NA) [2]. These multiplex techniques have been extensively developed and have three main categories. The first technique extended the two-dimensional (2-D) to three-dimensional (3-D) storage by the two-photon polymerization to induce the change of the refractive index in a voxel. Its data storage densities can exceed 1012 bits/cm3 as photorefractive materials Cibatool, photoresist [3], spirobenzopyran [4] and mixture of TNF and DMNPAA [5] are used. The stored data can be readout by measuring the changes of refractive index using differential interference contrast (DIC) microscopy. The second technique is exploiting wavelength for multiplexing by absorption modulation. Two or three wavelengths storage can be achieved by focusing multiple wavelength lasers onto the photochromic materials such as azobenzene molecules [6] and diarylethene [7] to induce the wavelength dependent absorbance change. The multiple wavelengths stored data can be readout by measuring the change of transmittance or reflectance using the same wavelength laser. The other technique to achieve multiplexing is to use polarization encoding by two-photon induced birefringence property via photoisomerization [8]. The multiple polarization state can be recorded by using different polarized laser to induce the different rotation of the optical axis of the material. The stored polarization state can be readout by the polarized laser scanning confocal microscopy [9]. All these techniques require materials to meet the corresponding mechanisms to increase dimensions, which makes it difficult to integrate two techniques into one material. However, in 2009, gold nanorods was used to encode across three wavelengths and two polarization states [10] to realize five-dimensional optical recording. This is the first time that two approaches have been integrated into a single technique which could ultimately increase the information capacity by orders of magnitude. The development of multi-dimensional optical data storage has brought substantial increase in optical data storage capacity. However, regardless of the DIC and reflectance measurement, the stored information can only be readout bit by bit.
On the other hand, the use of long and short codes and the integrating of multiple data recording layers has greatly increased the storage capacity of Blu-ray Discs [11]. But its reading throughputs are only about 20 MB/s [12], which are limited by the single-bit reading. To improve the reading throughputs of optical data storage, the technique of parallel processing using diffractive optical elements [13,14] to produce a 4×4 grid [13] has been proposed which divides the laser beam into multiple sub-laser beams to read multiple bits at the same time. Meanwhile, the parallel processing can also be achieved by micro lens arrays [15,16], dynamic holograms [17] and multifocal arrays [18]. However, such methods increase the reading complexity and the storage capacity is also restricted by the spacing of sub-laser spots.
To overcome the reading throughputs limit, we present an interference-aided reflectance spectral measurement to readout multidimensional optical data at one exposure. In order to readout the multidimensional data, disc with dielectric multilayer is designed by introducing an interference layer onto the substrate. In a storage unit, pits of different depths are written by laser direct writing to represent different bit of data. Because the different wavelengths have the different optical path lengths, with the help of the interference layer, a small change in the depth is enough to generate a difference of the transmittance or reflectance spectra. Such difference of spectra can be used to readout the stored multidimensional information simultaneously. This reading method is suitable not only for optical data storage by changing the optical path length, including the changes of refractive index and thickness of the storage medium, but also for super resolution optical data storage, which the size and spacing of the recorded spots are smaller than the diffraction limit.
This paper is structured as follows. Disc structure and principle of readout are introduced in the section two. The section three introduces the preparation of samples and experimental setup. In the section four, we analyze the peak shift of the reflectance spectra of different codes represented by combination of pits of different depths and demonstrate the feasibility of this readout method for multidimensional optical data.
2. Disc structure and principle of readout
Figure 1(a) shows the structure of the disc with dielectric multilayer in this study. A SiO2 substrate of 1 mm is utilized. SiO2 and Ta2O5 layers are deposited by the Leybold ARES1110 High Vacuum Coating System onto the substrate alternately to constitute the disc. In this structure, because of the wavelength dependence of the refractive indices of SiO2 and Ta2O5, the interference spectra are sensitive to the layer thickness. We simulate the reflectance and transmittance of the light through the optical disc by the transfer-matrix method. We optimize the thickness of layer structure to let the interference effect largest when the alternative layer thickness is comparable to λ/4 or thicker, where λ is the designed wavelength. In our simulation, we set the designed wavelength 630 nm. The optimized disc structure is as follows: SiO2 535 nm/ Ta2O5 146 nm/ SiO2 107 nm/ Ta2O5 73 nm/ SiO2 107 nm/ Ta2O5 73 nm/ SiO2 substrate.
We use photoresist as the recording layer deposited by spin-coating. The optical data is recorded onto this layer by laser direct writing. In our proof-of-concept demonstration, we use the combination of three different depths of pits in a storage unit to represent eight distinct information codes, which can be regarded as three bits of information representing 000, 001, 010, 011, 100, 101, 110, and 111 respectively. As shown in Fig. 1(b), it is encoded as 111. The pits where the depth information is written represents binary code 1, otherwise it represents 0. Different depths represent the 1 in different positions. Consequently, in the dielectric multilayer optical disc, the combination of three different depths of pits plays a role of a fundamental multidimensional storage unit which carry 3 bits information.
During the reading process, reading light focuses on a storage unit and covers all the pits in the unit. When the white light reflected from the unit with pits of different depths, the reflectance spectra imply the thickness information. The coherent white light is used in this paper and the coherent addition of the reflectance spectra of all pits in a storage unit is given by
Where ${R_n}$ is the reflectance spectra of the nth pit in a storage unit. ${k_n}$ is the proportion of area of the nth pit in the spot size. ${R_{n + 1}}$ represents the reflectance spectra of the area that has no pits but is covered by light and ${k_{n + 1}}$ is the proportion of this area in the spot size. In addition, the incoherent white light can also be used to read data because the peak position shift is mainly caused by self-interference. However, it is difficult to focus incoherent to small spot to match the recorded pits. In this paper, coherent light is used as the readout light for experiments which results in greater intensity contrast in the coherent reflectance spectra.3. Preparation and experimental setup
The RZJ-304 positive photoresist made by Suzhou Ruihong Electronic Chemicals Co. LTD is used as the recording layer. Positive photoresist is chosen here due to it is possible to engrave pits with a suitable depth by controlling the exposure time of laser direct writing. In order to design suitable pit depths, the accuracy of the thickness measurement for the dielectric multilayer optical disc needs to be determined in advance. In this paper, polymethyl methacrylate (PMMA)/anisole sol with 5% PMMA mass fraction is used. By changing the rotation speed of spin coating, PMMA film with different thickness are prepared on the disc.
The experimental arrangement as shown in Fig. 2 is employed to investigate the performance of the dielectric multilayer optical disc and the readout method of the multidimensional optical data. The 800 nm, 1KHz femtosecond light is focused on the BK7 glass to generate supercontinuum white-light [19] which is used to be the readout light. A filter is used to filter 800 nm femtosecond pump pulses to ensure that only supercontinuum white-light enters the readout system. Another white light generated by the halogen light source is used for monitoring and locating the recorded pits. These two beams of light are first focused onto the optical disc through an objective lens (50X NA0.5), and then the reflected light from the disc passes through the objective lens again and is imaged onto iHR320 Spectrometer or the camera. The camera is used to ensure the readout light focuses on the recorded pits. The stored optical information can be decoded by comparing the reflectance spectra measured by the iHR320 Spectrometer with the spectra calculated by formula (1).
Fig. 2. Schematic diagram of the interference-aided reflectance spectroscopy measurement to readout optical data. L1, L2, L3, L4, L5: convex lens; BS1, BS2: beam splitters; M1, M2: mirror; M3: flip mirror; Filter: 720 nm shortpass filter.
4. Results and discussion
The method of the interference-aided reflectance spectral measurement was proposed to determine the film thickness whose accuracy has been proven even less than 1 nm [20]. We use PMMA film as the undetermined film to verify this method. Sample 1 and sample 2 are prepared by spin coating with the rotation speed of 4000 rpm and 5000 rpm. Figure 3(a) shows the reflectance spectra of the disc without the recording layer. The black curve is the experimental reflectance spectra and the red one is the simulated result. The experimental results can fit the corresponding simulated results very well as expected. Figure 3(b) and Fig. 3(c) show the peak I and II in the reflectance spectra of sample 1 and sample 2 whose thicknesses are determined to be 142 nm and 134 nm, respectively. Atomic force microscope, ellipsometer and confocal microscopy are used to verify the accuracy of the interference-aided method. Table 1 lists the results of these four methods and the difference between three other measurements and the interference-aided method. The maximum difference is ±14 nm. The average thickness of the sample 1 and sample 2 measured by these four methods are 143 nm and 134.8 nm respectively. The difference between these two averages is 8.2 nm which is same as the thickness difference measured by the interference-aided method.
Fig. 3. (a) Both experimental and simulated reflectance spectra of the disc without PMMA film. The black and red curves represent the experimental (EXP) and simulated (SIM) reflectance spectra, respectively. (b)Peak I and (c)peak II of the experimental and simulated reflectance spectra of sample 1 (SPL1) and 2 (SPL2). (d) Simulated reflectance spectra of the disc with 100 nm to 140 nm PMMA films. (e) Simulated reflectance spectra of the disc with 130 nm to 140 nm PMMA films in peak II. (f) Experimental reflectance spectra of the disc with 134 nm and 136 nm PMMA film in peak II.
Table 1. Results of PMMA film thickness experimentally determined with different methodsa
The disc with dielectric multilayer is capable of measuring the thickness of the film more accurately. In Fig. 3(b) and Fig. 3(c), as a result of the PMMA film thickness is reduced by 8 nm, peak I and peak II move 3 nm and 4 nm to the short wavelength direction, respectively. The simulated reflectance spectra of the disc with 100 nm to 140 nm PMMA layer are shown in Fig. 3(d). By increasing 8 nm thickness of PMMA film, peak I, peak II and peak III move to the long wavelength direction by 3 nm, 4 nm and 4.6 nm respectively, which leads the large enough difference in reflected spectra. Meanwhile, when the thickness of undetermined film increases, the peaks of the reflectance spectra shift to the longer wavelength, and the shift of the peak at the long wavelength is greater than short wavelength. Figure 3(e) shows simulated reflectance spectra of the disc with 130 nm to 140 nm PMMA films in peak II. The peak shift of 1 nm can still be observed with only 2 nm thickness change of PMMA film. The experimental reflectance spectra of the disc with 134 nm and 136 nm PMMA film also have the peak shift of 1 nm as shown in Fig. 3(f). Therefore, our designed interference-aided optical disc has the ability to measure the 2 nm thickness difference of the undetermined film.
Owing to the high sensitivity to film thickness, the disc with dielectric multilayer can be used to readout multiple bits of stored data at one exposure. In conventional data storage devices, a storage unit accommodates only one bit of information (0 or 1). In our disc with dielectric multilayer, more than one bit of data is stored in a storage unit, represented by different depths recorded on the photoresist. The multiple bits of data can be readout by monitoring the peaks shift of the reflectance spectra caused by the interference. In order to make the reflectance spectra generate distinguishable peak shifts, various combinations of pits with different depths are utilized. 2-bit and 3-bit pit combination are designed as Fig. 4(a)(b) show. In the experiment, 650 nm photoresist is used as the recorded layer. We set the exposure time to 2s, 5s and 10s represented by dark blue, green and red to engrave pits with distinguishable depth. Two bits pit combination consists of pits with exposure time of 5s and 10s. Three bits pit combination consists of pits with exposure time of 2s, 5s and 10s. The images scanned by AFM are shown in Fig. 4(c)(d). In the figures, the diameter of each recorded pit is about 1um, and the distance between two adjacent pits in the same storage unit is 1.2um. We record the unit 5 times and the average depths of these pits are about 200 nm, 390 nm and 510 nm respectively. It is found that even each time the exposure time of laser direct writing are the same, the depth of the pits we engraved have fluctuations as shown the error bars in Fig. 4(e). These three pits which represent different binary codes, can be distinguished by different depths. In this experiment, pits are engraved by 80 MHz, 442 nm, and 40μW femtosecond laser and a 100X NA 0.8 objective lens.
Fig. 4. Schematic diagram of (a)2-bit and (b)3-bit pit combinations. The exposure time of dark blue, green and red pits is 2s, 5s and 10s, respectively. (c) 2-bit and (d) 3-bit pit combinations scanned by AFM. The scale bar in the figures represents 1um. (e)The depths of these pits exposed 2s, 5s and 10s.
2-bit pit combinations are tested first. Two pits are used to represent one bit, in order to increase the area of these pits and enhance interference shift. By matching the reflectance spectra with the simulated results, the two bits recording information can be read out. Due to the larger peak shift in long wavelength, only the peak III near 650 nm is used. The distribution of the peak positions of the experiment is shown in Fig. 5(a). X-axis represents the wavelength and Y-axis represents the four different binary codes. The graph of each color represents the peak position distribution of the reflection spectra of the corresponding binary code. Due to the fluctuation of pit depth in 5 recording, there is a peak distribution of the reflectance spectra of the same code. The peak distribution wave bands of the reflectance spectra of these four codes do not overlap. Therefore, the two bits stored data can be read out at once with low bit error rate. Figure 5(b)-(e) list the Gaussian fitting of the measured reflectance spectra of two bits pit combination for one of 5 recording and the simulated results of these pit combinations near 650 nm, of which peak positions match well. The simultaneous readout of 2-bit optical data storage can be accomplished.
Fig. 5. (a) Peak positions of Code 00 to 11. (b)-(e) Reflectance spectra of experimental and simulative results of Code 00, 01, 10, 11 at peak position near 650 nm. The black and red curves represent the experimental and simulated reflectance spectra, respectively.
In the experiment of 3-bit pit combination, each bit of data is represented by one pit as a result of the limitation of the reading spot size. In order to distinguish these eight codes at one exposure, two peaks in the reflectance spectra are used, which are around 655 nm and 520 nm respectively.
The peak positions distribution of the experiment is shown in Fig. 6(a). X-axis represents the wavelength and Y-axis represents eight different binary codes. Two wave bands of the same color on the same line represent peak positions distribution of the corresponding coded reflectance spectra. The selected two peak distribution wave bands of any two codes will not overlap at the same time. Figure 6(b)(c) list the reflectance spectra of Code 100 and 110 after Gaussian fitting and the simulated results of these pit combination near 520 nm. Figure 6(d)(e) list the reflectance spectra of Code 100 and 010 after Gaussian fitting and the simulated results of these pit combinations near 655 nm. Two peaks of experiment and simulation can match well. In other words, the data stored in the disc can be readout by the interference-aided spectra-fitting method, which is consistent with the prediction. A change in the depths of the photoresist can produce an obvious interference difference in the reflectance spectra, resulting in the fast readout function in multidimensional optical data storage. Due to the error in the laser direct writing, the peak position distribution of the reflectance spectra has a different wave band length, such as Code 10 in Fig. 5(a) and Code 111 in Fig. 6(a). Considering the error, peak position can be also distinguished clearly. Two or three bits are read at one time, which is equivalent to double or triple the reading rate of optical data storage. Meanwhile, if four bits are written in a storage unit, the reading speed is increased by four times.
Fig. 6. (a) Peak positions of Code 000 to 111. (b)(c) Reflectance spectra of experimental and simulative results of Code 100 and 110 at peak position near 520 nm. (d)(e) Reflectance spectra of experimental and simulative results of Code 100 and 010 at peak position near 655 nm. The black and red curves represent the experimental and simulated reflectance spectra, respectively.
5. Conclusion
In this study, we have developed a fast multidimensional optical data storage readout method based on interference-aided reflectance spectroscopy measurement. The multidimensional optical data represented by pits of different depths is recorded in the photoresist layer on the disc with dielectric multilayer substrate by laser direct writing. We extract 2 pits of two different depths or 3 pits of three different depths to form two or three bits of information which can be recorded as {(00), (01), (10), (11)} or {(000), (001), (010), (011), (100), (101), (110), (111)}. The stored data is read out by matching the measured reflectance spectra and the simulated results. The improvement of the readout ability is demonstrated by both simulations and experiments. The proposed method can readout two bits or three bits information at one exposure, which increases the reading rate of Blu-ray discs by two or three times respectively. Moreover, this reading method can be used for any optical data storage written by changing the optical path length and is easily to be used in the Blu-ray drive. For super-resolution optical storage with larger data capacity, our reading method also has a good application prospect.
Funding
Science and Technology Commission of Shanghai Municipality (18DZ1100403, 20DZ2210300).
Acknowledgment
We thank the staff members of the Molecular Imaging System at the National Facility for Protein Science in Shanghai (NFPS), Zhangjiang Lab, China for providing technical support and assistance in data collection and analysis.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. M Gu, X Li, and Y. Cao, “Optical storage arrays: a perspective for future big data storage,” Light Sci Appl 3(5), e177 (2014). [CrossRef]
2. K Kobayashi and S. Kano S, “Multi- Layered Optical Storage with Nonlinear Read/Write,” Opt. Rev. 2(1), 20–23 (1995). [CrossRef]
3. J H Strickler and W W. Webb, “Three-dimensional optical data storage in refractive media by two-photon point excitation,” Opt. Lett. 16(22), 1780 (1991). [CrossRef]
4. S., Kawata and Y. Kawata, “Three-dimensional optical data storage using photochromic materials,” Chem. Rev. 100(5), 1777–1788 (2000). [CrossRef]
5. D. Day, M. Gu, and A. Smallridge, “Rewritable 3D bit optical data storage in a PMMA-based photorefractive polymer,” Adv. Mater. 13(12-13), 1005–1007 (2001). [CrossRef]
6. T Gao, Y Xue, Z Zhang, and W. Que, “Multi-wavelength optical data processing and recording based on azo-dyes doped organic-inorganic hybrid film,” Opt. Express 26(4), 4309 (2018). [CrossRef]
7. S Pu, F Zhang, and J. Xu, “Photochromic diarylethenes for three-wavelength optical memory,” Mater. Lett. 60(4), 485–489 (2006). [CrossRef]
8. 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]
9. M Maeda, H Ishitobi, Z Sekkat, and S. Kawata, “Polarization storage by nonlinear orientational hole burning in azo dye-containing polymer films,” Appl. Phys. Lett. 85(3), 351–353 (2004). [CrossRef]
10. P Zijlstra, J Chon, and M. Gu, “Five-dimensional optical recording mediated by surface plasmons in gold nanorods,” Nature 459(7245), 410–413 (2009). [CrossRef]
11. Sergi Morais, Rosa Puchades, and Angel. Maquieira, “Disc-based microarrays: principles and analytical applications,” Anal. Bioanal. Chem. 408, 4523–4534 (2016). [CrossRef]
12. O Trelles, P Prins, M Snir, and RC. Jansen, “Big data, but are we ready?” Nat. Rev. Genet. 12(3), 224 (2011). [CrossRef]
13. L Sacconi, E Froner, R Antolini, MR Taghizadeh, A Choudhury, and FS. Pavone, “Multiphoton multifocal microscopy exploiting a diffractive optical element,” Opt. Lett. 28(20), 1918–1920 (2003). [CrossRef]
14. Y. Kuroiwa, N. Takeshima, Y. Narita, S. Tanaka, and K. Hirao, “Arbitrary micropatterning method in femtosecond laser microprocessing using diffractive optical elements,” Opt. Express 12(9), 1908–1915 (2004). [CrossRef]
15. JI Kato, N Takeyasu, Y Adachi, HB Sun, and S. Kawata, “Multiple-spot parallel processing for laser micronanofabrication,” Appl. Phys. Lett. 86(4), 044102 (2005). [CrossRef]
16. S. Hasegawa, Y . Hayasaki, and N. Nishida, “Holographic femtosecond laser processing with multiplexed phase Fresnel lenses,” Opt. Lett. 105(11), 1705–1707 (2006). [CrossRef]
17. A Jesacher and MJ. Booth, “Parallel direct laser writing in three dimensions with spatially dependent aberration correction,” Opt. Express 18(20), 21090–21099 (2010). [CrossRef]
18. H Lin, B Jia, and M. Gu, “Dynamic generation of Debye diffraction-limited multifocal arrays for direct laser printing nanofabrication,” Opt. Lett. 36(3), 406–408 (2011). [CrossRef]
19. J Wang, Y Zhang, H Shen, Y Jiang, and Z. Wang, “Spectral stability of supercontinuum generation in condensed mediums,” Opt. Eng. 56(7), 076107 (2017). [CrossRef]
20. X Liu, S Wang, H Xia, X Zhang, R Ji, T Li, and W. Lu, “Interference-aided spectrum-fitting method for accurate film thickness determination,” Chin. Opt. Lett. 14(8), 44–48 (2016). [CrossRef]
