We study the infrared photosensitivity properties of two optical polymer materials, benzocyclobutene (BCB) and epoxy OPTOCAST 3505, with a 10.6 μm CO2 laser. We discover that the CO2 laser radiation can lower the refractive index of BCB by as much as 5.5 × 10−3, while inducing no measurable index change in the epoxy. As confirmed by Fourier transform infrared spectroscopy, the observed index change in BCB can be attributed to photothermal modification of chemical bonds in the material by the CO2 laser radiation. Our findings open up a new possibility of processing polymer materials with a CO2 laser, which could be further developed for application in the areas of post-processing and direct-writing of polymer waveguide devices.
© 2011 OSA
Polymer optical waveguides have attracted much attention because of their potential for mass production of low-cost, high-performance devices . As the refractive index of polymer can in general be changed by ultraviolet (UV) radiation, UV writing has become a popular technique for the fabrication of polymer waveguide devices [1–7]. Compared with traditional microfabrication techniques, which involve complex processing steps and expensive equipment, laser writing can be much simpler and cheaper to implement. The present study explores the possibility of changing the refractive index of polymer by infrared radiation in order to take advantage of the availability of low-cost CO2 lasers.
CO2 lasers, which produce infrared radiation at 10.6 μm, are well-developed lasers with many industrial applications. A high power CO2 laser is frequently used for cutting and welding, while lower powered devices are used for engraving. Polymer [8, 9] and silica  waveguides have been fabricated by CO2 laser ablation, where the material in the non-guiding area is removed by the laser radiation. Recently, practically the same technique has been demonstrated for the fabrication of corrugated long-period grating filters in polymer waveguides . CO2 laser ablation, which is a pure photothermal effect, occurs at an energy density above a threshold . There is, however, no detailed report of any optical property modifications of polymer induced by CO2 laser radiation at an energy density below the ablation threshold. In our study, we measure the index changes of benzocyclobutene (BCB) (Dow Chemical Co.) and epoxy OPTOCAST 3505 (Electronic Materials Inc.) exposed to low-power CO2 laser radiation. BCB is a desirable waveguide material for its many nice properties, such as a high glass-transition temperature, excellent chemical and moisture resistance, and low optical loss , and the epoxy, which has a lower refractive index, works well as a cladding material for BCB waveguides . We discover that CO2 laser radiation can induce a large negative index change in BCB films without causing any visible damage to the material. An analysis by Fourier-transform infrared (FTIR) spectroscopy indicates modifications of chemical bonds in the material by the laser radiation. On the other hand, we observe no index change in epoxy films with similar dosage of CO2 laser radiation, which suggests that the infrared photosensitivity of polymer is strongly material-dependent. Our findings point to a new direction of processing polymer materials with a CO2 laser, which could find many applications in the development of polymer optical waveguide devices.
2. Thin film preparation
The samples used in our study were BCB and epoxy film waveguides fabricated on silica-on-silicon (SiO2/Si) substrates, where the SiO2 layer was 3 μm thick. The process of making a BCB film waveguide involved spin-coating of a BCB film onto a substrate and curing of the film in a nitrogen (N2) atmosphere . We prepared two BCB samples, which differed from each other in the oxygen (O2) level present in the gas used in the curing process. For the “N2 cured BCB”, curing took place in a 99.99% pure N2 atmosphere with a temperature ramp up to 270 °C for ~60 min, where the O2 level was negligible (< 100 ppm). Under this curing condition, the BCB film was >95% cured (polymerized), according to the information provided by the supplier . For the “O2 cured BCB”, the film was cured in a < 99% pure N2 atmosphere, where the O2 level was much larger than 100 ppm. In the O2 cured BCB, the cyclobutene reactant group or the tetrahydronaphthalene polymerization product was oxidized . We prepared an epoxy film with the same spin-coating technique, where the epoxy film was first cured in air by exposure to the radiation from a UV lamp (Novacure 2100) at an intensity of 350 mW/cm2 for 3 min and then cured thermally at 130 °C for 60 min.
We measured the refractive index and the thickness of the polymer films with a prism-coupler system (Metricon 2010) and a surface profiler (Ambios XP-2), respectively. The refractive indexes of the N2 cured BCB sample, measured at 1536 nm, were 1.5348 and 1.5317 for the transverse-electric (TE) and transverse-magnetic (TM) polarizations, respectively, while the corresponding indexes of the O2 cured BCB sample were 1.5576 and 1.5537, which is consistent with the supplier’s information that curing in the O2 environment can raise the refractive index . On the other hand, the refractive indexes of the epoxy film at 1536 nm were 1.5113 and 1.5112 for the TE and TM polarizations, respectively. The two BCB films were 2.6 μm thick and the epoxy film was 6.6 μm thick. All three samples were 3 cm wide and 4 cm long. The propagation losses of the N2 cured and O2 cured BCB samples, also measured with the prism coupler at 1536 nm, were ~1 and ~3 dB/cm, respectively. The oxidation bonds formed in the curing process resulted in a larger optical loss in the O2 cured BCB film. In our experiments, each of these samples was cut into a number of smaller samples, which were irradiated with different energy densities of CO2 laser radiation.
3. CO2 laser induced refractive index change in BCB films
The CO2 laser used in our study (CO2-H10, Han’s Laser) had a maximum average output power of 10 W and was pulse-width modulated at a frequency of 5 kHz. The laser beam was focused to a spot of ~100 μm in diameter and controlled with a computer to scan across the waveguide surface. The laser beam first scanned across the waveguide in the transverse direction in 1 μm steps at a speed of 50 mm/s and then advanced along the waveguide in the longitudinal direction by 1 μm to start a new scan in the transverse direction. After a sufficient area of the waveguide was scanned, the scan was repeated, but this time stepping along the longitudinal direction first and then the transverse direction. These two rounds of scans across the waveguide constituted a scan cycle. The laser energy density was controlled by adjusting the average output power of the laser (i.e., the pulse width). This CO2 laser has been used for writing long-period gratings in different types of optical fibers [14, 15].
We first present the results for O2 cured BCB samples. Each sample was scanned by the CO2 laser beam. After each scan cycle, the refractive index and the film thickness were measured by the prism-coupler system and the surface profiler, respectively. The dependence of the index change on the number of scan cycles for 5 samples irradiated at different CO2 laser energy densities is shown in Fig. 1 for both the TE and TM polarizations, where the measurement uncertainties are determined by the index resolution (± 5 × 10−4) of the prism-coupler system. As shown in Fig. 1, the index change decreases rapidly at the beginning and becomes saturated. The effects of CO2 laser irradiation on BCB are distinctly different from those of UV irradiation, which induces a positive index change in BCB . In the case of CO2 laser irradiation, the saturated index changes which are almost the same for both polarizations, depend sensitively on the laser energy density. The maximum index change achieved at the energy density 180 mJ/mm2 (−5.5×10−3) is five times larger than that at 100 mJ/mm2 (−1.1×10−3). In the case of UV irradiation, a higher UV energy density can only shorten the time to reach saturation; it cannot increase the saturation value . The thicknesses of the samples did not change after CO2 laser irradiation. The microscopic images of the sample before and after 10 scan cycles of irradiation at 180 mJ/mm2 are shown in Figs. 2(a) and (b) , respectively. No damage on the sample surface caused by the CO2 laser radiation can be seen. When the energy density was lower than 100 mJ/mm2, no index change was induced, regardless of the number of scan cycles used. When the energy density was somewhat higher than 180 mJ/mm2, ablation was observed. The microscopic image of a sample after one scan cycle of irradiation at 200 mJ/mm2 is shown in Fig. 2(c), where serious damage on the film surface can be seen.
We repeated the measurements for the irradiated samples after 6 months and obtained practically the same results, which confirms that the CO2 laser induced index change in the BCB films are permanent and stable at room temperature.
We next present the results for the N2 cured BCB samples. The index changes measured at 100 and 180 mJ/mm2 of CO2 laser radiation are shown in Figs. 3(a) and (b) , respectively. The index change at 100 mJ/mm2 is almost negligible, while the maximum index change at 180 mJ/mm2 is ~−1.0 × 10−3, which is much smaller than that observed with the O2 cured BCB sample irradiated at the same energy density. The film thickness did not change after laser irradiation. The microscopic images of a N2 cured BCB film before and after 6 scan cycles of CO2 laser irradiation at 180 mJ/mm2 are shown in Figs. 4(a) and (b) , respectively. No damage caused by the laser radiation can be seen. Damage due to ablation became serious at 200 mJ/mm2, even though only one scan cycle was used, as shown in Fig. 4(c).
To investigate any possible chemical modifications in the BCB films due to laser irradiation, we analyzed the BCB samples with FTIR spectroscopy . Figure 5 shows the infrared absorption spectra of the BCB film measured before and after the second scan cycle of CO2 laser irradiation at 180 mJ/mm2, as well as their difference. The laser radiation causes a decrease in the absorption bands at around 1775 and 3040 cm−1, which are associated with the absorption peaks at 1640 – 1780 and 2500 – 3300 cm−1 for the carbonyl (C = O) and hydroxyl (O–H) groups, respectively . Our results agree with those reported for CO2 laser ablated polyimide , low-temperature-resistant epoxy resin , and polyethylene terephthalate . In those cases, a decrease in the absorption bands for specific groups can also be observed, which indicates the breaking or elimination of specific chemical bonds. In UV irradiated BCB, however, the absorption for these two groups increases . As UV radiation has a large photon energy (≧4 ev) and is a good source for photochemical processing of organic systems , the increase in the absorption indicates the formation of the C = O and O–H groups in BCB after UV irradiation [7, 20]. The increase in the refractive index of the BCB film after UV irradiation can therefore be attributed to the formation of the oxidization bands . Compared with UV radiation, CO2 laser radiation has a much smaller photon energy (0.176 eV) and can directly excite the vibrational modes in organic solids. The excitation energy is rapidly converted to bulk heating. Therefore, the effect of CO2 laser irradiation is mostly photothermal . The observed decrease in refractive index after CO2 laser irradiation should be due to the breaking or elimination of the oxidization bands formed during the curing process in the presence of O2. It is known that, when the BCB film is processed in an O2 environment, an absorption band near 1775 cm−1 is produced as a result of oxidization and the dielectric constant of the film in the oxidization band can be increased by up to 10% . In the N2 cured BCB samples, fewer oxidization groups are formed during the curing process, which means fewer C = O and O–H groups can be broken or eliminated by laser irradiation. This explains why the index changes in N2 cured BCB samples after CO2 laser irradiation are much smaller. In addition, a larger CO2 laser energy density can induce a higher temperature in the polymer , which may result in further elimination of C = O and O–H groups and hence a larger saturation index change.
During the refractive-index measurements with the prism coupler, we observed no obvious changes in the widths of the attenuated total reflection peaks for those exposed samples that showed no visible surface ablation, which implies little changes in the optical loss. More careful experiments are needed to measure any losses induced by CO2 laser exposure.
4. CO2 laser induced refractive index change in epoxy films
We repeated the experiments for the epoxy film with different energy densities of CO2 laser radiation and found no measurable induced index change. The infrared absorption spectra of the epoxy film measured before and after CO2 laser irradiation at 150 mJ/mm2 are almost the same, as shown in Fig. 6 , which indicates that CO2 laser irradiation does not cause any significant chemical modifications in the epoxy film. The chemical analysis from FTIR spectroscopy is consistent with the index measurement result. The microscopic images of the epoxy film before and after 3 scan cycles of CO2 laser irradiation at 100 mJ/mm2 are shown in Figs. 7(a) and (b) , respectively. Weak ablation was observed after irradiation at 100 mJ/mm2, as shown in Fig. 7(b). Ablation became serious after one scan cycle of irradiation at 150 mJ/mm2, as shown in Fig. 7(c), where the grid-like ablation pattern is due to nonuniform scanning of the CO2 laser beam across the film surface. These results indicate that the damage threshold of the epoxy film is highly sensitive to the laser energy density.
We discover that CO2 laser radiation at 10.6 μm can lower the refractive index of a BCB film cured in the O2 environment by as much as 5.5 × 10−3 without causing any visible damage to the film. The index change is permanent and stable at room temperature. On the other hand, the same amount of CO2 laser radiation induces a much smaller index change to a BCB film cured in a pure N2 environment and practically no index change in an epoxy OPTOCAST 3505 film. The infrared photosensitivity of polymer is highly material-dependent and could be further explored for application in the areas of post-processing and direct-writing of polymer waveguide devices.
The work described in this paper was fully supported by a research grant from City University of Hong Kong [Project Nos. CityU 7002444].
References and links
1. L. Eldada and L. W. Shacklette, “Advances in polymer integrated optics,” IEEE J. Sel. Top. Quantum Electron. 6(1), 54–68 (2000). [CrossRef]
2. L. Eldada, C. Xu, K. M. T. Stengel, L. W. Shacklette, and J. T. Yardley, “Laser-fabricated low-loss single-mode raised-rib waveguiding devices in polymers,” J. Lightwave Technol. 14(7), 1704–1713 (1996). [CrossRef]
3. J. S. Koo, R. B. Williams, C. B. E. Gawith, S. P. Watts, G. D. Emmerson, V. Albanis, P. G. R. Smith, and M. C. Grossel, “UV written waveguide devices using crosslinkable PMMA-based copolymers,” Electron. Lett. 39(4), 394–395 (2003). [CrossRef]
4. L. Y. Chen, W. S. Tsai, W. H. Hsu, K. Y. Chen, and W. S. Wang, “Fabrication and characterization of benzocyclobutene optical waveguides by UV pulsed-laser illumination,” IEEE J. Quantum Electron. 43(4), 303–310 (2007). [CrossRef]
5. W. C. Wang, M. Fisher, A. Yacoubian, and J. Menders, “Phase-shifted Bragg grating filters in polymer waveguides,” IEEE Photon. Technol. Lett. 15(4), 548–550 (2003). [CrossRef]
6. D. G. Rabus, P. Henzi, and J. Mohr, “Photonic integrated circuits by DUV-induced modification of polymers,” IEEE Photon. Technol. Lett. 17(3), 591–593 (2005). [CrossRef]
7. K. P. Lor, K. S. Chiang, Q. Liu, and H. P. Chan, “Ultraviolet writing of buried waveguide devices in epoxy-coated benzocyclobutene,” Opt. Eng. 48(4), 044601 (2009). [CrossRef]
8. C. Dinger, T. Sterkenburgh, T. Holler, and H. Franke, “Patterning multimode polymeric lightguides using a CO2 laser,” Proc. SPIE 1774, 278–287 (1993). [CrossRef]
9. S. S. Zakariyah, P. P. Conway, D. A. Hutt, D. R. Selviah, K. Wang, H. Baghsiahi, J. Rygate, J. Calver, and W. Kandulski, “Polymer optical waveguide fabrication using laser ablation,” in Proceedings of IEEE Conference on Electronics Packaging Technology (Institute of Electrical and Electronics Engineers, New York, 2009), 936–941.
10. L. Ç. Özcan, F. Guay, R. Kashyap, and L. Martinu, “Investigation of refractive index modifications in CW CO2 laser written planar optical waveguides,” Opt. Commun. 281(14), 3686–3690 (2008). [CrossRef]
11. Q. Liu and K. S. Chiang, “CO2-laser writing of polymer long-period waveguide gratings,” in Proceedings of IEEE Conference on Photonics Global Singapore (Institute of Electrical and Electronics Engineers, New York, 2008), C62–C65.
12. F. Kane and R. R. Krchnavek, “Benzocyclobutene optical waveguides,” IEEE Photon. Technol. Lett. 7(5), 535–537 (1995). [CrossRef]
13. Dow Chemical Company, Processing Procedures, http://www.dow.com/cyclotene/prod/302235.htm.
14. Y. Liu, K. S. Chiang, Y. J. Rao, Z. L. Ran, and T. Zhu, “Light coupling between two parallel CO2-laser written long-period fiber gratings,” Opt. Express 15(26), 17645–17651 (2007). [CrossRef] [PubMed]
16. B. H. Stuart, Infrared Spectroscopy: Fundamentals, and Applications (Wiley, Chichester, 2004).
17. D. W. Zeng and K. C. Yung, “XPS investigation on Upilex-S polyimide ablated by pulse TEA CO2 laser,” Appl. Surf. Sci. 180(3-4), 280–285 (2001). [CrossRef]
18. A. Hartwig, G. Vitr, S. Dieckhoff, and O.-D. Hennemann, “Surface treatment of an epoxy resin by CO2 laser irradiation,” Die Angew. Makromol. Chem. 238(1), 177–189 (1996). [CrossRef]
19. M. Dadsetan, H. Miradeh, and N. Sharifi, “Effect of CO2 laser radiation on the surface properties of polyethylene terephthalate,” Radiat. Phys. Chem. 56(5-6), 597–604 (1999). [CrossRef]
20. W. W. Duley and R. E. Mueller, “CO2 laser welding of polymers,” Polym. Eng. Sci. 32(9), 582–585 (1992). [CrossRef]