Open Access
Add to BibSonomyAbstract
We describe a procedure for rapidly and conveniently prototyping a periodic structure at submicrometer order using holographic interferometry and micro-molding processes. In this experiment, the master of the periodic structure was created on an i-line submicrometer positive photoresist film by a holographic interference using a He-Cd (325nm) laser. A subsequent mold using polydimethylsiloxane (PDMS) polymer was cast against this master and used as a stamp to transfer the grating pattern onto a UV cure epoxy. The technique shows accurate control for the transferring of a grating’s period and depth. The grating pattern on the epoxy produced by the PDMS mold shows an average of less than 2% error in the grating period and an average of 15% error in depth reproduction.
©2005 Optical Society of America
1. Introduction
Gratings used in integrated optics have been recognized as very important elements for many applications such as wavelength filtering, sensing, and optical measuring techniques [1–7]. Polymer gratings are widely investigated because of their low cost and simple fabrication. Typical techniques for patterning gratings on polymers include holographic lithography [8–10], electron-beam (e-beam) lithography [11], laser-beam direct writing [12], x-ray mask technology [13], and phase mask lithography [14–15]. Although an e-beam lithography process provides flexibility and high resolution for defining almost any prescribed pattern, it is limited to very small throughput and has difficulty obtaining accurate pattern placement. Alternatively, laser-beam direct writing techniques provide the same flexibility for pattern definitions, but the etching profile is highly dependent on the operating wavelength. Thermal effects stemming from laser ablation also cause unwanted irradiated material that collects around the edges of the etched material. The advantages of the holographic interferometric techniques are (1) that they can provide flexibility in grating period and depth by changing the incident beam angles and (2) irradiation time. In addition, the theoretical limit of the frequency of the interference pattern produced by two intersecting beams is half of the wavelength of the incident beam alone. Thus, the grating period is limited only by the wavelength of the light source. The materials used as well as the fabrication processes are important factors in manufacturing optical elements for different applications. The sol-gel hybrid (SGH) materials have shown simple fabrication of grating diffraction by the holographic interferometric technique, but this material cannot be fabricated to obtain a high aspect ratio of the grating pattern [16]. Therefore, our polymer diffraction gratings were fabricated using the holographic interferometric technique with a photoresist (Ultra 123) to obtain high aspect ratio grating patterns as the first step for our three-step grating manufacturing process.
There are many replication processes that are simple and involve fairly easy fabrication, such as hot embossing [17], UV-embossing [18], and the micro-transfer molding method [19–20]. Although these methods are ideal for mass production, they still have some problems to overcome, such as lip problems or the limitation of just using substrate and core material [17–20]. However, if these molding processes are combined with a LIGA-like process that can produce a mold for the subsequent molding processes, it shows great mass production potential [21]. Utilizing a similar concept, this paper will describe a technique for combining the holographic, interferometric, and micromolding processes to create a high resolution grating structure on a polymer surface.
2. Methodology
The rapid prototyping of grating structures on a polymer substrate involves a three-step process. First, a grating pattern is holographically exposed using a two-beam interference pattern on a positive photoresist film. This produces a master that can be subsequently used to produce a polydimethylsiloxane (PDMS) mold. The silicone rubber mold then is used as a stamp to transfer the final gratings pattern onto a UV cure epoxy polymer. The following sections describe the process involved for the grating fabrication.
The master grating patterns were holographically exposed using a two-beam interferometer technique (Fig. 1). The light source is a 60 mW, 325 nm wavelength single mode (TEM00) Helium-Cadmium laser (Kimmom Electric LTD Corp.). The grating period formed by the interference pattern is determined by
where T is the grating period, λ is the wavelength of laser light, and θ1 and θ2 are the incident angles of the two plane waves coming to the surface plane, respectively. In order to reduce the sensitivity of the grating period change, which is caused by changing the two incident angles, we fix the angle θ2 and just adjust the angle θ1 to obtain the required grating periods. From Eq. (1), one can easily derive that
where ΔT is the grating period deviation caused by the angle θ1 change, Δθ1. From Eq. (2), the grating period can be accurately controlled by a single reflecting angle θ 1. For example, if the angle θ2 is fixed at 12° and then the angle θ1 is adjusted to 26.237°, the grating period T equals 500 nm. If a 10 nm period change (ΔT = 10 nm) is required, it can be done by rotating the angle θ1 by 0.83°. The angle resolution is 0.001°.
The holographic interference pattern was exposed on an i-line submicrometer positive photoresist (Ultra 123 made by MicroChem Corp., Mass., refractive index = 1.618). A precut 1.0 by 2.0 cm2 Pyrex glass substrate was spun-coated with a layer of 0.6 μm thick Ultra 123 photoresist. This sample was first baked in a convection oven at 90°C for 120 s to remove excess solvent from the photo-resist. The sample was then placed on the sample holder for the holographic light exposure (Fig. 1). In this experiment, the angle θ 2 was fixed to 12° and the angle θ 1 was adjusted to obtain the desired grating periods. Three groups of the grating periods, 500, 600, and 700 nm, were fabricated. These samples were post-baked in an oven at 110°C for 120 s. After baking, the sample was rinsed in MF26A developer (MicroChem Corp., Mass.) for 30 s until the grating pattern was obtained. The overall exposed dimensions of the gratings were 3 mm long and 1 mm wide.
One of the control parameters is the grating depth. Here a relation between grating depth and the exposure time was examined (Fig. 2). Based on the experiment, a 5 min exposure gave a grating depth of 430 nm and 492 nm for the 500 nm and 700 nm grating periods, respectively. Maximum exposing time for the process is around 5.5 min. The equivalent dosage is around 1.65 mW/mm2. Above the maximum exposure time, the resist becomes overexposed and no pattern is formed on the film.
Fig. 2. Grating depth as a function of light exposing time on photoresist. (The combined intensity of two incident beams is 15μW for all experiments.)
The profiles of the grating were measured by an atomic force microscope (AFM) using a high aspect ratio silicon based tip (AR5T made by NANOSENSORS, Switzerland). The tip is designed with a tip radius less than 15 nm and a tip half cone angle smaller than 2.8°. The high aspect ratio portion of the tip is longer than 1.5μm. The length to diameter aspect ratio is greater than 10:1. At a tip height of 500nm, the corresponding tip diameter is only 57nm. With this tip diameter, it should not affect the measurement of our groove depth (~500nm). The tip is also designed with an additional 13° tilt angle to the center axis of the tip to provide compensation caused by the mount of the atomic force microscope so that the tip will stand exactly perpendicular to the sample surface. This feature enables users to measure deep and narrow structures as well as very steep sidewalls with very little distortion. However, for the grating design, the depth and the duty cycle measurements are usually sufficient to determine grating strength and coupling efficiency. Therefore the profile measurement is not as critical in the grating design.
For example, in an exposure time of 3 min for a grating period of 500 nm, the AFM result showed the grating period was 505 nm and the grating depth was 333 nm [Fig. 3(a)]. For the case of a 5 min exposure time on a 700 nm grating structure, the AFM result showed a grating period of 703 nm and grating depth of 492 nm [Fig. 3(b)]. Based on the preliminary results, we found that the grating period and the corresponding depth of the grating pattern can be accurately controlled down to an error rate of less than 1%. We also found that a high aspect ratio of almost 1:1 between the depth and the period of the grating structure could be obtained using this process.
The patterned resist was used as a master mold to transfer the grating pattern onto a polydimethylsiloxane (PDMS) thin film using a typical micro-molding technique [22–23]. The PDMS was first diluted in a hexane solution before spin-coating on the patterned Ultra 123 photoresist. After baking at 90°C for 1 hr, the PDMS was cured and easily peeled off from the Ultra 123 mold (Fig. 5). The profiles of the PDMS molds were measured using an atomic force microscope (AFM). Based on the results from the AFM, when the depths of the gratings were less than 350 nm, grating patterns transferred quite accurately from the positive photoresist. The AFM micrograph result shows that the gratings on a PDMS substrate transferred from a master grating, a profile of 490 nm period and 186 nm deep, are slightly off from the original dimension (Fig. 4). The grating period was off by 4.1nm while the depth was reduced by 9.8 nm.
Fig. 3. The AFM picture and measurement result for the grating on photoresist (a) 500 nm grating period and (b) 700 nm grating period.
A table comparing the grating geometry on the PDMS and photoresist molds shows that the overall dimension was reduced when patterns were transferred from the photoresist to PDMS (Table 1). Based on the results, the reduction shows an average of 0.22% or 1.4 nm average reductions in the periods. The depths, on the other hand, were reduced by as much as 12.68% and 20 to 30 nm, on average.
An interesting fact is observed when the aspect ratio between the depth and the period is greater than 0.7, where some of the gratings appeared to affix to the bottom and each other after release from the photoresist mold. This effect was observed when the depth of the grating was larger than 350 nm and the grating period was smaller than 500 nm. The same effect was also observed when the depth and period were greater than 450 nm and 650 nm, respectively. Under the scanning electron microscope (SEM), these structures appeared as though the grating period had been broadened (Fig. 6). Upon a more careful inspection, we found the sticking was due to the fact that when the aspect ratio between the depth and period is close to 0.7, a non-uniformly distributed free charge built up on the PDMS surface after it was peeled from the photoresisit, which created enough electrostatic force to overcome the support and cause the fins structure to collapse. Based on the measurement taken from the electroscope, the negative charge on the surface of the PDMS was about 1.5 × 10-10 ~ 4 × 10-10 Coulombs over an area of 1 × 1 cm2 PDMS. It is likely that a weakly supported fins structure, due to the elastic nature of the PDMS, was overcome by the electrostatic force generated from the negative charges on the PDMS surface. This sticking effect increased when the aspect ratio increased because the fins were less rigid. However, this effect disappeared once the ratio fell below 0.7.
Fig. 6. The SEM micrograph of gratings on PDMS (500 nm grating period). Some fins are stuck together as shown in the figure.
The final pattern was transferred onto a UV cure epoxy substrate from the PDMS mold using an UV-replication process (Fig. 5) [24]. The fabrication procedure is described as follows. A spacer with a thickness of 400 μm was placed between the mold and a thin Pyrex glass slide. The mold was supported by another Pyrex glass slide to create a support for the PDMS mold. After injection of the precure UV polymer (OG146, EPOXY TECHNOLOGY Inc.), which has a relatively low viscosity (82 cps), into the opening between the mold and the glass slide using a fine tip syringe, the liquid solution automatically spread and filled up the space between the mold and slide due to capillary effects. A UV curing lamp with a wavelength range of 300–400 nm was used to crosslink the polymer at an intensity of 100 mW/cm2 for 1 to 2 min. After the polymer was fully cured, the polymer was easily peeled off from the mold, and the final gratings were formed on the polymer. In order to avoid the period irregularity caused by the sticking effect of the PDMS mold, which was mentioned earlier, the depths of the PDMS molds were restricted to features less than 300 nm, while the grating periods were varied from 460 to 700nm.
The results of from the SEM and AFM measurements on photoresist, PDMS, and UV polymer gratings are listed in Table 1. A SEM micrograph shows the top view of the UV polymer grating with.a 492 nm period, indicating that the gratings pattern was successful in transferring from the PDMS mold (Fig. 7). Overall, the depth reduction averaged around 15% from the original photoresist mold. However, the average error for the depth is less than 5% when the pattern is transferred from PDMS to UV polymer. On average, the period transferred much better where an average of 0.5% reduction occurred in both pattern transferring from photoresist to PDMS and PDMS to UV polymer. We can conclude that when the grating period is less than 600 nm, the period on the final UV polymer has an average of 1% reduction. However, when the period is greater than 600 nm, the reduction appears to be higher. Based on the quality of the grating, this process shows great potential for mass production of any period of grating structure.
3. Conclusion
We have successfully created a process to rapidly produce submicrometer range gratings by using both micro-molding and holographic interference techniques. A large aspect ratio of close to 1:1 ratio on the grating pattern can be obtained and consistent reproduction of the grating on a UV polymer could be produced. The grating period and depth on the UV gratings exhibited small differences from the originally designed grating pattern. This process shows great potential for mass production of any period of grating structure.
Table 1. The results of gratings from the SEM and AFM measurement on photoresist (PR), PDMS, and UV polymers.
References and Notes
1. N. Imoto, “An analysis for contradirectional-coupler-type optical grating filters,” J. Lightwave Technol. 3, 895–900 (1985). [CrossRef]
2. H. Hillmer, A. Grabmaier, H. L. Zhu, S. Hansmann, and H. Burkhard, “Continuously chirped DFB gratings by specially bent waveguides for tunable lasers,” J. Lightwave Technol. 13, 1905–1912 (1995). [CrossRef]
3. C. H. Lin, Z. H. Zhu, Y. Qian, and Y. H. Lo, “Cascade self-induced holography: a new grating fabrication technology for DFB/DBR lasers and WDM laser arrays,” IEEE J. Quantum Electron 32, 1752–1759 (1996). [CrossRef]
4. Y. Shibata, S. Oku, Y. Kondo, and T. Tamamura, “Effect of sidelobe on demultiplexing characteristics of a grating-folded directional coupler demultiplexer,” IEEE Photonics Technol. Lett. 8, 87–89 (1996). [CrossRef]
5. A. Sharon, D. Rosenblatt, and A. A. Friesem, “Narrow spectral bandwidths with grating waveguide structures,” Appl. Phys. Lett. 69, 4154–4156 (1996). [CrossRef]
6. S. Yin, F. T. S. Yu, and S. Wu, “Optical monitoring for plasma-etching depth process,” IEEE Photonics Technol. Lett. 4, 894–896 (1992). [CrossRef]
7. W.C. Wang, W.C. Chuang, C.T. Ho, Y.R. Lian, C.K. Chao, and R.F. Shyu,“A new novel means of transducing tensile stress,” in Proceedings of Progress in Electromagnetics Research Symposium, Z.A. Kong, ed. (Hangzhou, China, 2005), pp.89–91.
8. D. Y. Kim, S. K. tripathy, L. Li, and J. Kumar, “Laser-induced holographic surface relief gratings on nonlinear optical polymer films,” Appl. Phys. Lett. 66, 1166–1168 (1995). [CrossRef]
9. J. W. Kang, M. J. Kim, J. P. Kim, S. J. Yoo, J. S. Lee, D. Y. Kim, and J. J. Kim, “Polymeric wavelength filters fabricated using holographic surface relief gratings on azobenzene-containing polymer films,” Appl. Phys. Lett. 82, 3823–3825 (2003). [CrossRef]
10. S. Aramaki, G. Assanto, G. I. Stegeman, and M. Marciniak, “Realization of integrated Bragg reflectors in DANs-polymer waveguides,” J. Lightwave Technol. 11, 1189–1195 (1993). [CrossRef]
11. H. Nishihara, Y. Handa, T. Suhara, and J. Koyama, “Electron-beam directly written micro gratings for integrated optical circuits,” in Photo- and Electro-Optics in Range Instrumentation, J. Water, et al., eds., Proc. SPIE , 134, 152–159 (1980).
12. C. Y. Chao, C. Y. Chen, C. W. Liu, Y. Chang, and C. C. Yang, “Direct writing of silicon gratings with highly coherent ultraviolet laser,” Appl. Phys. Lett. 71, 2442–2444 (1997). [CrossRef]
13. N. Mukherjee, B.J. Eapen, D.M. Keicher, S.Q. Luong, and A. Mukherjee, “Distributed Bragg reflection in integrated waveguides of polymethylmethacrylate”, Appl. Phys. Lett. 67,3715–3717 (1995). [CrossRef]
14. L. Eldada, S. Yin, C. Poga, C. Glass, R. Blomquist, and R.A. Norwood, “Integrated multichannel OADMS using polymer Bragg grating MZIS,” IEEE, Photonics Technol. Lett. 10, 1416–1418 (1998). [CrossRef]
15. K. O. Hill, B. Malo, D. Bilodeau, D. C. Johnson, and J. Albert, “Bragg gratings fabricated in monomode photosensitive optical fiber by UV exposure through a phase mask,” Appl. Phys. Lett. 62, 1035–1037 (1993). [CrossRef]
16. D. J. Kang, J. K. Kim, and B. S. Bae, “Simple fabrication of diffraction gratings by two beam interference method in highly photosensitivity hybrid sol-gel films,” Opt. Express 12, 3947–3953 (2004), http://www.opticsexpress.org/abstract.cfm?URI=OPEX-12-17-3947. [CrossRef] [PubMed]
17. Holger Becker and Wolfram Dietz, “Microfluidic devices for TAS applications fabricated by polymer hot embossing,” in Microfluid Devices and Systems, A. B. Frazier and C. H. Ahn, eds., Proc. SPIE 3515, 177–181 (1998). [CrossRef]
18. P. M. Ferm and L. W. Shackjette, “High volume manufacturing of polymer waveguides via UV- Embossing,” in Linear, Nonlinear, and Power-Limiting Organics, E. Manfred, et al., eds., Proc. SPIE 4106, 1–10 (2000). [CrossRef]
19. K. E. Paul, T. L. Breen, J. Aizenberg, and G. M. Whitesides, “Maskless Photolithography: embossed photoresister as its own optical element,” Appl. Phys. Lett. 73, 2893–2895 (1998). [CrossRef]
20. X.-M. Zhao, S. P-Smith, S. J. Waldman, G. M. Whitesides, and M. Prentiss, “Demonstration of waveguide couplers fabricated using microtransfer molding,” Appl. Phys. Lett. 71, 1017–1019 (1997). [CrossRef]
21. H. D. Bauer, W. Ehrfeld, M. Harder, T. Paatzsch, M. Popp, and F. Smaglinski, “Polymer waveguide devices with passive pigtailing: an application of LIGA technology,” Synth. Metals 115, 13–20 (2000). [CrossRef]
22. J. C. Lotters, W. Olthuis, P. H. Veltink, and P. Bergveld, “The mechanical properties of the rubber elastic polymer polydimethylsiloxane for sensor applications,” J. Micromech. Microeng. 7, 145–147(1997). [CrossRef]
23. P. Nussbaum, I. Philipoussis, A. Huser, and H. P. Herzig, “Simple technique for replication of micro-optical elements,” Opt. Eng. 37, 1804–1808 (1998). [CrossRef]
24. M. Rossi, H. Rudmanr, B. Marty, and A. Maciossek, “Wafer-scale micro-optics replication technology,” in Lithographic and Micromaching Techniques for Optical Component Fabrication II, E.-B. Kley and H.P. Herzid, eds., Proc. SPIE 5183, 148–154 (2003). [CrossRef]
