This work designed and fabricated an optical pickup system based on optical films using the roll-to-roll process. The design combined the advantages of the stacked and planar optical pickup system. Two blazed gratings were used as beam splitters for bending the optical path, while a cylindrical lens was used for astigmatic focus-error detection. The proposed design effectively reduces overall system configuration, component cost, and fabrication complexity.
© 2011 OSA
Optical storage devices have become a rapidly growing market in recent years due to the high demands from mobile information technology (IT) products. Optical pickup plays an important role in optical storage devices. An optical pickup is a device that regenerates data by collecting reflected laser light from the disc, or writes data on optical disc by laser light. For compact disc (CD)/digital versatile disc (DVD)/ blue-ray disc (BD) usage, the wavelength is 780 nm, 650 nm, and 405 nm, respectively, and the numerical aperture (NA) value for objective lens is 0.45, 0.6, and 0.85, respectively. Besides, the radius of spot size must be smaller than the airy disk ( = 0.61 μm). A conventional optical pickup system is typically assembled from several components, including laser diode (LD) as a light source, collimation lens, beam splitter, astigmatic lens, quarter-wave plate (QWP), and an objective lens. Current optical pickup systems work sufficiently well for practical usage; however, they encounter issues such as large configuration, complex assembly, and costly alignment problems. Therefore, there is still room for improvement. To reduce assembly cost and produce smaller systems, various types of integrated optical pickup systems based on some form of semiconductor fabrication processes have been proposed. These include free space micro-bench [1,2], stacked type [3–5], and planar optics [6,7]. Diffractive optical elements (DOEs) are widely used for various functions such as optical components within a pickup system . These include fabrication methods for DOE's. Using photolithography to fabricate highly efficient DOEs requires several alignment steps, which easily suffer serious noise due to light scattering. Another issue is the high precision requirements to fabricate objective lens which is over the processability of photolithography approaches. A more promising high precision method is to use a focused ion beam (FIB) to fabricate a DOE . Although a FIB can produce micro-blazed structures for DOEs, it suffers from high cost and time-consuming manufacturing and is rarely used for commercial applications. A more reliable process is the roll-to-roll film technique, demonstrated in this work with blazed grating and efficiency over 75% applied in a novel color separation system . This process can produce a sub-micrometer optical profile with nanometer-level roughness. It holds a considerable potential for mass production of highly efficient DOEs for other optical systems as well.
In this study, a polymer-based optical pickup which transmission of light can reach 92%, as high as glass, was designed. And it could be thinner and lighter than glass-based optical pickup. The system composed of two types of blazed gratings, one refractive cylindrical lens, and an objective lens. The blazed gratings are designed to diffract and split the laser beam while the cylindrical lens is for astigmatic focus-error detection. This system combines the advantages of stacked type and planar optical pickup systems. Planar optics is a suitable integration platform for opto-electronic components, and can be realized in 3-D topology with 2-D complexity. For staked type, it increases optical efficiency usage for the transmissive DOEs. The proposed system used the transmissive DOEs to make them integrated on one bulk to realize 3-D topology. Except for the objective lens, all optical components were fabricated using the roll-to-roll imprinting process. By applying roll-to-roll fabricated optical films to realize optical pickup is a novel idea, the proposed optical pickups have highly potential to be mass production.
2. System design and simulation
This optical pickup is a finite-conjugate system for DVD applications. The system has an image numerical aperture (NA) of 0.6 in accordance with the adopted laser wavelength of 635 nm and the objective lens. To realize the function of a beam splitter in a conventional pickup, two blazed gratings were used to separate the forward and backward optical paths. We designed three types of optical pickup systems, consisting of a laser diode (LD), blazed gratings, a cylindrical lens, and an objective lens. In these systems, blazed gratings were applied to bend the optical path to reduce the volume of the whole configuration. Figure 1(a) shows the first type of designed optical pickup system. The laser beam emitting from the LD is focused by the objective lens to read the signal on the disc. The reflected backward laser beam is diffracted by the Grating 2 (G2) into the −1st diffraction order with an angle of θ2. Then the deflected beam is deflected again into the + 1st diffraction order of the Grating 1 (G1) at an angle of θ1. Finally, the laser beam carrying the disc data information and disc distance from the objective lens is focused onto the photo detector by an astigmatic cylindrical lens. The second type (shown in Fig. 1(b)) is similar to the first type, but bends the beam at first and the deflected beam propagates along the optical axis. The last type of designed optical pickup system, shown in Fig. 1(c), uses Grating 3 (G3) to diffract the deflected beam into the −1st diffraction order with an angle of θ1.
2.1 Design of the blazed grating
The diffraction grating used to deflect light is a critical component in the proposed system. Compared with other types of grating, blazed grating exhibits higher diffraction efficiency in the deflection direction. This design adopted three one-dimensional transmission blazed gratings, G1, G2, and G3. Their extraction (main diffraction) direction θdiff depends on the wavelength λ, the incident angle θinc, and the grating period d, given by the grating equations:
Here, m is an integer number corresponding to the diffraction order, and n ( = 1.495) are the refractive indices in the incident and diffraction media, respectively. As illustrated in Fig. 2 , the maximum grating depth is denoted as h, the passive angle as β, and the base angle as α. Among these parameters, the base angle α of the main diffraction facet can be estimated in the simulation using Snell’s law:
In simulation, another parameter t represents the optimization result given by the relation:
To obtain the highest diffraction efficiency and reduce cross-talk between diffraction orders, the grating shapes of G1, G2, and G3 are optimized using the commercial software GSOLVER 5.1. The diffraction efficiency is defined to the ratio of the intensities of the desired diffracted beam and the illuminating beam. All G1, G2, and G3 have grating periods of 4 μm for process limitation. The shape of G1 is designed to produce maximum efficiency in the 1st diffraction order under the angle of zero incidence. The simulation showed an optimal depth h of G1 between 1.1 μm and 1.2 μm when the tG1/d is 0.25 as shown in Fig. 3 . The diffraction intensities for G2 in the 1st and 0th orders are set equal under an angle incidence of zero, according to the G2 depth determined to be 0.64 μm with tG2/d = 0.3695. The diffraction intensities for G3 in the 1st and −1th order are set equal under the angle incidence of zero, for maximum efficiency with the h = 0.9 μm and tG3 = 2 μm as shown in Fig. 3. A monocrystal diamond-cutting tool, with nanometric edge sharpness, to cut the copper (Cu) roller was used in the process of making the master for blazed gratings and cylindrical lens. However, repeated cutting will gradually blunt the diamond bite, which results in rounding of the peak of the blazed grating, and, in turn, give rise to a loss of diffraction efficiency. To determine the acceptable loss of diffraction efficiency and its corresponding range of profile deviation, the tolerances of the profiles of two blazed gratings have to be calculated before fabrication. As shown in Fig. 4 , the efficiencies of the 0th and 1st diffraction orders of G1, G2, and G3 are plotted versus the rounding radiuses R. Considering the process feasibility and system requirement, the radius R is limited to be smaller than 0.4μm for both G1 and G2 such that the diffraction losses remain less than 3%. The diffraction loss for G3 decreases by only 0.6% for R up to 0.7 μm.
2.2 Design of cylindrical lens and system simulation
The cylindrical lens is used to focus the reflected light from the disc onto a four-quarter photodiode, where the focus error signal (FES) is generated, which can be used to monitor the relative distance between the objective lens and the disc. Therefore, the parameters of the cylindrical lens are related to the specification of the objective lens. Considering the available cutting tools for fabricating the master, the curvature radius of the designed cylindrical lens was first set at 510 μm, with its thickness and period set at 220 μm and 642 μm, respectively, while the distance between the objective lens and cylindrical lens has been optimized to be 13.78 mm.
The proposed system used a photo detector to monitor the relative distance between the objective lens and the surface of the disc. The cylindrical lens was used to focus on a four-quarter photodiode to generate a focus error signal (FES). Disc vibration prevents the objective lens from exactly projecting the focus spot on the data surface. If the disc position is in front of the focal point of the objective lens, the FES is (A + C) - (B + D) < 0, otherwise, if the disc position is behind the focal point of the objective lens, the FES is (A + C) - (B + D) > 0. The geometrical focal spot size is typically equal to zero in perfect focusing. Figure 5 displays the S-curve which illustrates the relationship between the disc position and the FES variation. Figure 6(a)-(c) presents the specifications of the micro-optical pickup system simulated by ZEMAX. Figure 6(d) gives the simulation results of FES for three types of systems, and Table 1 shows the S-curve data.
To realize our micro-optical pickup system based on ultra precision-machined optical films, this study analyzes three systems, taking into account the light efficiency of the whole system, S-curve specifications, and feasibility of the fabrication process. The S-curve represents the relationship between the disc position and the FES variation. The linear range of the designed system is about 28 μm which can be used for detection in this proposed system. Moreover, when PMMA substrate is integrated to the system, the linear range can be reduced. The efficiency of Type 3 is lower than that of Type1 or Type 2 (refer to Section 2.1). Since Type 1 and Type 3 are difficult to integrate into one PMMA bulk, we adopted Type 2 to realize and verify our idea. Table 2 shows the lens design data.
3. Experiment results and discussion
Grating diffraction efficiency strongly depends on a number of factors such as grating period, geometry, and refractive index of materials; therefore, a reliable fabrication process is necessary. This study fabricates optical components based on ultra-precision machining, one of the most important techniques developed in manufacturing high precision components. The advantages of ultra-precision techniques include accurate machining and mass production capability. A roll-to-roll imprinting process was used to make the gratings and cylindrical lens. The diamond tool was first utilized to define the mechanically grooved structures on an imprinting Cu roller. The UV resin was then dispensed on a 188 µm thick PET film (T-A4300 by Toray Corp.), imprinted by the roller, hardened by the UV source. After a series of tests, we found the optimal condition with the feed rate, 5~6 m/min, and with the absorbing energy, 900~1100 mJ/cm2 from UV light. With the specifications, gratings was able to achieve a high replication rate (>95%) and high demolding yield. Finally, the cured UV resin was demolded from the roller, as shown in Fig. 7 .
Figures 8(a) (b) display the SEM images of G1 and G2; Fig. 9 shows a microscope photo of the cylindrical lens array. The adopted blazed angles for G1 were α = 76°, β = 14°, and (180°- α - β) = 90°. For G2, blazed angles were α = 60°, β = 10°, and (180°- α - β) = 110°. (The definition of α, β are illustrated in Fig. 2) The measured transmission efficiencies of G1 were η+1, 633nm = 65.5%, η0, 633nm = 14%, and its simulation values were η+1, 633nm = 70.3%, η 0, 633nm = 11.9%. For G2, the measured transmission efficiencies were η0, 633nm = 42.3%, η-1, 633nm = 32.8%, and its simulation values were η0, 633nm = 38.9%, η-1, 633nm = 38.2%. Figure 10 compares the transmission efficiencies of the measured and simulated values of G1, G2, and the intensity distribution of their corresponding CCD images. The deviation of measured values from the simulation was acceptable for the reflection loss at the interface. For G1, the deviation of measured values from the simulation was less than 3%. For G2, the measurement results deviated 5% from the simulation for both the 0th and −1st orders, caused by the depth error (R ~0.35µm) in the fabrication process. The target efficiency for G2 is 42% for both the −1st and 0th orders. The good control of depth of gratings influences the diffraction efficiency, therefore, the design procedure need to consider the factor of experimental error for compensation. The surface smoothness is a key parameter for gratings and cylindrical lens to determine their performance. In this paper, the surface roughness was measured using 3D Laser Scanning Microscope (VK-9700, KEYENCE Corp.). The surface roughness of cylindrical lens was Ra: 50~150 nm, for V-cut patterns (gratings), the roughness was around Ra: 50~80 nm according to our pretests. The surface roughness can be further improved by tuning process parameters if necessary.
Figure 11 illustrates the entire pickup module for measurement, including a laser diode, two objective lens, and three passive optical films. For simplicity, a pinhole was used to block the unwanted light and the setup was folded at the disc position. Hence, the other objective lens was added for this testing module. The spot diagrams on the photo diode, representing the relative distances between the objective lens and the disc, are shown in Fig. 12(a) -12(c). The elongation direction of spot diagrams varies significantly with the relative direction and distance. The measured images converted to FES are indicated as solid dots in Fig. 13 . Figure 14 shows the system module, which contains blazed gratings and cylindrical lens fabricated using the process in Fig. 7.
Applying optical structured films to realize conventional optical functions can greatly improve the cost of current optical pickups. This paper presents a new optical pickup based on two blazed gratings and a cylindrical lens. Except for the objective lens, all the optical components were fabricated using the roll-to-roll imprinting process to meet the optical precision requirement. This single process of creating a blazed structure is important for fabricating high-performance blazed gratings, to minimize optical aberrations without misaligning grating patterns. The demonstration of blazed gratings shows that the deviation between the measured and the theoretical diffraction efficiency is less than 5%. Also this study analyzed the tolerances for the profiles of blazed gratings to determine the acceptable loss of diffraction efficiency. By applying optical structured films to realize some conventional optical functions, the proposed optical pickups hold a high potential for mass production. With our design, the overall system configuration, component cost, and fabrication complexity can be significantly reduced.
This study was sponsored by the National Science Council of Republic of China, Taiwan under contract number NSC-99-2221-E-007-041-MY3.
References and links
1. L. Y. Lin, J. L. Shen, S. S. Lee, and M. C. Wu, “Realization of novel monolithic free-space optical disk pickup heads by surface micromachining,” Opt. Lett. 21(2), 155–157 (1996). [CrossRef]
4. H. F. Shih, C. L. Chang, J. Lee, and C. S. Chang, “Design of optical head with holographic optical element for small form factor drive systems,” IEEE Trans. Magn. 41(2), 1058–1060 (2005). [CrossRef]
5. C. H. Lee, Y. Chiu, and H. P. Shieh, “Astigmatic diffractive optical element for swing-arm-type optical pickup head,” Opt. Eng. 48(7), 075201 (2009). [CrossRef]
8. J. Turunen, and F. Wyrowski, “Diffractive Optics for Industry and Commercial Applications,” 1st.ed. (Berlin: Akademi. Verlag, 1997)
9. H. B. Kim, G. Hobler, A. Steiger, A. Lugstein, and E. Bertagnolli, “Simulation-based approach for the accurate fabrication of blazed grating structures by FIB,” Opt. Express 15(15), 9444–9449 (2007). [CrossRef]
10. H. H. Lin, C. H. Lee, and M. H. Lu, “Dye-less color filter fabricated by roll-to-roll imprinting for liquid crystal display applications,” Opt. Express 17(15), 12397–12406 (2009). [CrossRef]