A plastic waveguide laser doped with organic dye molecule was fabricated with self-written active (SWA) waveguide technique. The device has a Fabry-Perot resonator consisting of a pair of highly reflective dielectric mirrors, which has brought two advantages for efficient optical pumping; (i) the efficient optical feedback in the cavity can be induced, and (ii) the reflection band of the dielectric mirrors can be tuned to overlap only with the emission band of the doped dye. For the SWA waveguide devices, furthermore, the active waveguide core is essentially coupled with a fiber port for optical input. Owing to these advantages, an experimental configuration for the optical end pumping can be easily applied. The high absorption efficiency for the pumping light could be obtained in this pumping method. A remarkable lowering of the lasing threshold was observed. As the best results of this study, consequently, the lasing action under the optical pumping energy as low as 50 nJ was achieved.
© 2010 OSA
Plastic laser devices doped with active media have promise as a compact light source for novel microoptic systems such as the optical integrated circuits and the photonic chips for biosensing, gas-analyses, and so on [1–7]. The advantages of the plastic devices are, for example, ease in fabrication of the plastic films, consistency in the refractive index with peripheral silica components, and the large gain properties of the doped material such as organic dyes [1–9] and organic complexes with rare-earth ions [10,11]. In the past, various types of the plastic laser have been widely investigated. A popular type frequently used for demonstration of the lasing phenomena would be ‘planar’ waveguides with a cavity structure for the optical feedback. For example, the Fabry-Perot (FP) cavity has been obtained in a simple way, i.e. cleaving of the sample edges in the both sides [9,10,12], whereas the optical feedback is not so efficient due to the low index of the plastic material. The distributed feedback cavities have also been fabricated by several methods  such as using a substrate with a corrugated surface profile , making a gain grating with the two-beam interference [3,15–17], and fabrication of grating corrugation by a technique of nanoimprint lithography [4–6,8,18–20]. Toward the practical application, however, the ‘planar’ waveguide would be unsuitable for an integrated light source, because all of the peripheral optical components must have structures of the ‘channel’ waveguide. Since the laser emission should be efficiently coupled into the peripheral components, development of a plastic laser source having both a channel waveguide structure and an efficient optical feedback cavity would be strongly required [3,4].
Recently, we have developed self-written active (SWA) waveguide technique as a very simple method for the fabrication of a plastic waveguide laser [21–23]. The channel waveguide with a large optical gain can be self-formed in UV-curable resin doped with organic dye molecules . By the light irradiation from a waveguide port, the resin is polymerized with increase in the refractive index. Then the transversal confinement of the irradiated light is induced. Due to the chain-reaction occurrence of the polymerization, the increase in refractive index, and the light confinement, a fiber-type active waveguide core with a constant diameter can be obtained [22–28]. Most recently, furthermore, we have demonstrated that a FP resonator can be easily applied to the SWA waveguide . The fabrication of the SWA-FP device was also very simple; a pair of half-mirrors, which were silica glass slide (thickness ~150 mm) with vacuum-evaporated Al film (thickness ~100 nm), was preliminary set up between the tips of the two optical fibers. After that, in the same way as described above, the SWA waveguide was fabricated by the exposure from the two fibers. The FP lasing was clearly observed in the emission measurement under the optical pumping by a pulsed light. It was confirmed that the lasing threshold showed a reasonable variation against the length of the FP cavity . Nevertheless, the reflectivity of the Al-coated half-mirrors was as low as 0.2, which could not give an ‘efficient’ FP cavity. For the previous device, furthermore, the optical pumping was performed only from the side surface of the waveguide that is referred to as the side-pump configuration. The absorbance of the pump light becomes fairly low at the side-pump configuration because of the small length for interaction between the pumping photons and the active media. Moreover, the cylindrical profile of the side surface has been unfavorable for the side-pump configuration, because it causes a substantial reflection of the pump light. These facts mean the dissipation of the pumping photons, and result in a high lasing threshold for the total power of the incident light.
In this study, we have improved the design of the SWA-FP device and the method of optical pumping to achieve a considerable decrease in the lasing threshold. Instead of the Al-coated half-mirrors, the FP resonator of the improved device was configured by a pair of high-reflectivity mirrors, which brought two advantages to lower the lasing threshold. The first, of course, was that the optical loss at the reflection could be marginally reduced due to the nearly 100-% reflectivity at the emission band. Secondly, since the reflection band of the mirrors could be tuned to exclude the wavelength for the optical pumping, the optical pumping from the waveguide port, i.e. the end-pump configuration, could be easily applied. In addition, the waveguide core of the SWA-FP laser device was essentially coupled with the peripheral optical fibers, which was also helpful to apply the end-pump configuration. In this experimental scheme, the lasing threshold was consequently decreased to as low as 50 nJ.
2. Fabrication and measurements
The base material of the plastic waveguide laser was a copolymer of pentaerythritol triacrylate (PETA) and benzyl acrylate (BA). The active medium doped in the plastic waveguide was an organic dye molecule, LDS798, which has a maximum absorption at 566 nm and the lasing emission band of 770 – 830 nm . A mixture of the PETA and BA monomer resins doped with LDS798 and a photoinitiators was used as the precursor for the waveguide fabrication . The volume ratio of the PETA and BA monomers were 9: 1. The dye concentration in the copolymer was varied in a range of 0.1 – 0.6 wt%.
Figure 1(a) represents the layout for fabrication of the SWA-FP device. Two optical fibers were roughly aligned on a grooved silica substrate. The waveguide core of the optical fiber had a graded-index (GI) profile, and its diameter was 62.5 μm. Into the groove between the fiber tips, a pair of silica plates (thickness ~150 μm) was set up to make a FP resonator. A dielectric multilayer of SiO2 and TiO2 was deposited on the silica plates, which would act as a distributed Bragg reflection (DBR) mirror for the emission. After setting up the fibers and mirrors, the mixture of the PETA and BA monomers doped with LDS798 was cast into the groove. Then laser lights with the wavelength of 405 nm were introduced from the fiber tips to induce the self-formation of the SWA waveguide. Typical conditions for the bi-directional exposure were 20 μW and 30 s. Under these conditions, the SWA waveguide with the length over 1.5 mm can be fabricated. From a separate measurement for a sample with doping concentration of 0.3 wt%, the absorption coefficient at 405 nm was estimated to be 14 cm−1. This means that transmission of approximately 10% of incident power can be ensured at the waveguide length of 0.8 mm. The diameter was almost the same as the core diameter of the optical fibers (62.5 μm). This means that the SWA waveguide fabricated here has been a transversal multimode waveguide. The propagation loss of the SWA waveguide was estimated in a recent study to be ~0.54 mm−1 . As shown in Fig. 1(b), the reflection band of the DBR mirrors (reflectivity R > ~99.8%) was designed to be 740 – 840 nm. This wavelength range well overlaps with the emission band of LDS798, and does not include the wavelength for the exposure (405 nm). In the fabrication, therefore, the DBR mirrors allowed the good transmission of the exposed light, and thus formation of the SWA waveguide inside the cavity was not obstructed. Consequently, the two waveguides were coupled inside the cavity. The uncured resin was removed. For comparison, the devices, in which the Al-coated mirrors (R ~0.2) were used for the FP resonator, were also fabricated.
The optically pumped emission measurements were performed for the SWA-FP devices. The light source used for the optical pumping was a wavelength-tunable liquid dye laser system, in which an ethanol solution of Rhodamine B was excited by the third harmonic of pulsed lights from a Nd-doped yttrium aluminum garnet laser. The operation wavelength of the liquid dye laser was varied in a range of 520 – 610 nm. The width of the pumping pulse was ~1 ns. Neutral density filters were used for regulation of the pulse energy for pumping. The emission output from one of the optical fiber was detected by a spectrometer with an array of charge-coupled devices. The spectral resolution was approximately ~1 ns.
3. Results and discussions
Figure 2 shows a typical emission spectrum of the SWA-FP device. In this sample, the DBR mirrors were used for the FP resonator, and the cavity length L and the dye concentration n were 1.14 mm and 0.3 wt%, respectively. The measurement was carried out at the side-pump configuration at 610 nm. The beam profile of the pumping pulse from the light source was reshaped to be rectangular, so that the SWA waveguide in the cavity was irradiated uniformly. As shown in the figure, the spectrum has a number of sharp emission lines. This emission spectrum attributes to a FP lasing. The inset shows the emission intensity as a function of the pumping density of the pulsed light, Ipump. The lasing threshold can be evidently observed at Ipump = 0.2 mJ/cm2.
The pitch of the ripples observed in the emission spectrum is ~5 nm. On the other hand, the mode separation of the FP emission is expected to be ~0.2 nm, because of the cavity length as large as 1 mm. Therefore it is considered that the ripples do not directly show the longitudinal cavity modes. We speculate that such the spectral profile is caused by the ‘transversal’ multimode propagation. Since each transversal mode has an individual effective index, the wavelengths and separation of the longitudinal cavity modes are also individual for each. The observed emission should be a combination of these, which would be a very complicate profile with a spectral beat. In the actual measurement with a low spectral resolution, the envelope profile of the beat was considered to be observed.
Variation of the lasing threshold with L was investigated. As shown by closed circles in Fig. 3 , the lasing threshold of the samples with the DBR mirrors remained around 0.2 mJ/cm2 and showed no distinctive variation against L. On the other hand, the relatively high lasing thresholds were observed for the samples of the Al-coated half-mirrors (see closed triangles). Furthermore, a significant increase with decreased L was observed. The difference in the behavior of the lasing threshold can be easily understood with Eq. (1).
An additional benefit of using the DBR mirrors was high transmissivity at the wavelength for the exposure. Unlike the Al-coated half-mirrors, the DBR mirror provided the sufficient UV transparency (>~0.85 at 405 nm) together with the efficient optical feedback at the emission wavelength, as shown in Fig. 1(b). This feature was very effective to fabricate the samples with the higher doping concentration. We could fabricate the samples with n of 0.6 wt%, whereas n was up to 0.3 wt% in the device with the Al-coated mirrors. The further decreasing in the lasing threshold was observed in these devices, as shown by open circles in Fig. 3. As the best case, the lasing threshold of 22.3 μJ/cm2 was achieved at L = 0.9 mm. These results show that the drastic reduction in lasing threshold could be achieved for the SWA-FP laser device by introducing the high reflection mirrors as the FP resonator.
Next we will show the impact of the end-pump configuration on the lasing action of the SWA-FP device. As described in the introduction section, absorbance of the pump light is fairly small at the side-pump configuration. Additionally, light irradiation onto the side surface of the waveguide with a cylindrical profile causes a substantial reflection. These mean the dissipation of the incident photons and are fatal disadvantages for the efficient optical pumping. To induce the lasing action, consequently, the light source with the excessive output power is required for the optical pumping. A simple solution would be application of the end-pump configuration. Conveniently, the waveguide core of the SWA-FP device was essentially coupled with the peripheral optical fibers, which was a very favorable feature to apply the end-pump configuration. However, in the previous device design, in which the Al-coated mirrors were used as the FP cavity, the transmissivity at the pump wavelength was as low as ~0.2. Then the sufficient pumping power for lasing could not be provided. In the device with the DBR mirrors, on the other hand, the transmissivity as large as ~0.85 could be obtained at the wavelength region for the optical pumping, as shown in Fig. 1(b). Thus it was expected that most incident photons would be absorbed by the active media in the cavity, and could contribute to the formation of the population inversion.
Figure 4 shows emission spectra of the SWA-FP devices at the end-pump configuration. The pump wavelength was 610 nm, and the device had L = 1.24 mm cavity configured with the DBR mirrors. In this measurement, the dye concentration n was reduced to be ~0.1 wt% because we would like to ensure the sufficient penetration length of the pump light. With increasing the pulse energy for pumping, sharp emission peaks appeared, and a rapid increase in the emission intensity was observed, as shown in the figure. These observations reveal the FP lasing has been achieved at the end-pump configuration as well as the side-pump configuration.
A comparison of the lasing thresholds between the side-pump (closed circles) and end-pump (open circles) configurations is shown in Fig. 5 . The horizontal axis is Ppump, which is the pulse energy of the light source used for the optical pumping. For the data at the side-pump configuration, Ppump was defined as a product of Ipump, L ( = 1.24 mm), and the diameter of the waveguide (62.5 μm). This means that all of the power described as Ppump was used for the actual pumping of the device. The lasing threshold at the side-pump configuration was found to be ~400 nJ. In the end-pump configuration, a substantial reduction in the lasing threshold was confirmed; Ppump at the threshold was ~137 nJ. This result, of course, reveals that the absorption of the pumping photons is more efficient in the end-pump configuration, whereas the efficiency at the side-pump configuration was so small because of the waveguide core with a much smaller diameter than the absorption length. It was found in a rough estimation that almost 100% of the pulse energy would be absorbed at the end-pump configuration whereas less than ~37% at the side-pump configuration. This estimation can account for the result shown in Fig. 5; the ‘absorbed’ pulse energy could be expected to be comparable at the two configurations.
Figure 6(a) shows a variation of the lasing threshold with the pump wavelength λpump. The data for the sample of L = 1.24 mm, which was used in Figs. 4 and 5, are shown by the closed squares. The lasing threshold showed a significant variation with the pump wavelength, and reached ~430 nJ at λpump = 520 nm, whereas the threshold as small as 137 nJ had been obtained at 610 nm. The triangles and the circles show the data for the samples of L = 1.01 and 0.72 mm, respectively. One can find that the shorter L also gives the significant reduction in the lasing threshold. Consequently, the best result of this study was the threshold as low as 48 nJ, which was obtained in the sample of L = 0.72 mm under the optical pumping at 595 nm.
Let us discuss the mechanism of the threshold varied with the pump wavelength. As shown in Fig. 6(b), LDS798 has maximum absorption at ~570 nm . While the maximum absorption was expected to induce the population inversion easily, the lower threshold was found at the longer wavelength, as shown in Fig, 6(a). This wavelength shift seems to be attributed to a wavelength dependence of the penetration depth for the pump light. When the absorption coefficient, which means a product of the Molar absorptivity and the volume density of the doped dye, is so large, the penetration of the pump light would be inhibited, and the waveguide length with a positive gain could not be extended. This phenomenon would be a competing effect with the stimulation of a large optical gain by the strong absorption. The trade-off relationship appears rather clearly in the samples with L = 1.01 and 0.72 mm, in which the distinct local minimum can be observed at the long-wavelength side of the absorption maximum (590 – 600 nm). On the other hand, such the local minimum was not found at the short-wavelength side of the absorption maximum. While this discrepancy is still an open question, the existence of the nonradiative deactivation process from the higher excited states is considered as one of the possible candidates.
We have fabricated an organic dye-doped plastic laser by using a very simple method, self-written active waveguide technique. This device has a FP resonator consisting of a pair of DBR mirrors. For the efficient optical pumping of this device, the end-pump configuration was applied. The SWA-FP device has the several advantages for this pumping method; (i) the active waveguide core is essentially coupled with a fiber port for the optical input, (ii) the reflection band of the mirrors can be tuned to overlap with the emission band and not to include the wavelength for optical pumping, and (iii) the high-reflectivity at the emission band causes the efficient optical feedback in the cavity. The lasing threshold was drastically reduced from the previous result, in which a FP resonator was fabricated with Al-coated half-mirrors. As the best result of this study, the lasing threshold as low as 50 nJ was obtained. Furthermore, it was demonstrated that the low lasing threshold of the SWA-FP device remains even at the very short cavity length. From theses results, the SWA-FP device has promise as integrated light source for the optical integrated circuit and the photonic chip devices.
This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan, Nos. 21750192 and 22550163. This work was supported in part by the Research Grant from the Ogasawara Foundation for the Promotion of Science & Engineering.
References and links
2. A. Costela, I. García-Moreno, and R. Sastre, “Polymeric solid-state dye laser: Recent developments,” Phys. Chem. Chem. Phys. 5, 4745–4763 (2003) (and references therein). [CrossRef]
3. Y. Oki, Y. Ogawa, K. Yamashita, M. Miyazaki, and M. Maeda, “Integration of optical pumped dye laser on organic microflowcytometry chip,” Mol. Cryst. Liq. Cryst. (Phila. Pa.) 463(1), 131–140 (2007). [CrossRef]
5. C. Ge, M. Lu, W. Zhang, and B. T. Cunningham, “Distributed feedback laser biosensor incorporating a titanium dioxide nanorod surface,” Appl. Phys. Lett. 96(16), 163702 (2010). [CrossRef]
7. J. Schmidtke and E. M. Terentjev, “Polydimethylsiloxane-enclosed liquid crystal lasers for lab-on-a-chip applications,” Appl. Phys. Lett. 96(15), 151111 (2010). [CrossRef]
8. K. Yamashita, M. Arimatsu, M. Takayama, K. Oe, and H. Yanagi, “Simple fabrication technique of distributed-feedback polymer laser by direct photonanoimprint lithography,” Appl. Phys. Lett. 92(24), 243306 (2008). [CrossRef]
9. S. Yuyama, T. Nakajima, K. Yamashita, and K. Oe, “Solid state organic laser emission at 970 nm from dye-doped fluorinated-polyimide planar waveguides,” Appl. Phys. Lett. 93(2), 023306 (2008). [CrossRef]
11. K. Yamashita, H. Taniguchi, S. Yuyama, K. Oe, J. Sun, and H. Mataki, “Continuous-wave stimulated emission and optical amplification in europium (III)-aluminum nanocluster-doped polymeric waveguide,” Appl. Phys. Lett. 91(8), 081115 (2007). [CrossRef]
12. V. G. Kozlov, V. Bulovic, P. E. Burrows, and S. R. Forrest, “Laser action in organic semiconductor waveguide and double-heterostructure devices,” Nature 389(6649), 362–364 (1997). [CrossRef]
13. H. Kogelnik and C. V. Shank, “Stimulated emission in a periodic structure,” Appl. Phys. Lett. 18(4), 152–154 (1971). [CrossRef]
14. D. Schneider, U. Lemmer, W. Kowalsky, and T. Riedl, “Low-threshold organic semiconductor lasers” in Organic Light Emitting Devices, K. Müllen and U. Scherf ed. (Wiley-VCH, Weinheim, 2006), p. 369–395, and references therein.
15. M. Fukuda and K. Mito, “Solid-state dye laser with photo-induced distributed feedback,” Jpn. J. Appl. Phys. 39(Part 1, No. 10), 5859–5863 (2000). [CrossRef]
16. F. Sobel, D. Gindre, J.-M. Nunzi, C. Denis, V. Dumarcher, C. Fiorini-Debuisschert, K. P. Kretsch, and L. Rocha, “Multimode distributed feedback laser emission in a dye-doped optically pumped polymer thin-film,” Opt. Mater. 27(2), 199–201 (2004). [CrossRef]
17. N. Tsutsumi, A. Fujihara, and D. Hayashi, “Tunable distributed feedback lasing with a threshold in the nanojoule range in an organic guest-host polymeric waveguide,” Appl. Opt. 45(22), 5748–5751 (2006). [CrossRef] [PubMed]
18. M. Berggren, A. Dodabalapur, R. E. Slusher, A. Timko, and O. Nalamsu, “Organic solid-state laser with imprinted gratings on plastic substrate,” Appl. Phys. Lett. 72(4), 410–411 (1998). [CrossRef]
19. J. R. Lawrence, G. A. Turnbull, and I. D. W. Samuel, “Polymer laser fabricated by a simple micromolding process,” Appl. Phys. Lett. 82(23), 4023–4025 (2003). [CrossRef]
20. Y. Chen, Z. Li, Z. Zhang, D. Psaltis, and A. Scherer, “Nanoimprinted circular grating distributed feedback dye laser,” Appl. Phys. Lett. 91(5), 051109 (2007). [CrossRef]
21. K. Yamashita, E. Fukuzawa, A. Kitanobou, and K. Oe, “Self-written active waveguide for integrated optical amplifier,” Appl. Phys. Lett. 92(5), 051102 (2008). [CrossRef]
24. M. Kagami, T. Yamashita, and H. Ito, “Light-induced self-written three-dimensional optical waveguide,” Appl. Phys. Lett. 79(8), 1079 (2001). [CrossRef]
25. S. Shoji, S. Kawata, A. A. Sukhorukov, and Y. S. Kivshar, “Self-written waveguides in photopolymerizable resins,” Opt. Lett. 27(3), 185–187 (2002). [CrossRef]
26. O. Sugihara, H. Tsuchie, H. Endo, N. Okamoto, T. Yamashita, M. Kagami, and T. Kaino, “Light-induced self-written polymeric optical waveguides for single-mode propagation and for optical interconnections,” IEEE Photon. Technol. Lett. 16(3), 804–806 (2004). [CrossRef]
27. K. Yamashita, T. Kuro, K. Oe, K. Mune, T. Hikita, and A. Mochizuki, “Propagation-mode-controlled fabrication of self-written waveguide in photosensitive polyimide for single-mode operation,” IEEE Photon. Technol. Lett. 17(4), 786–788 (2005). [CrossRef]
28. K. Saravanamuttu and M. P. Andrews, “Visible laser self-focusing in hybrid glass planar waveguides,” Opt. Lett. 27(15), 1342–1344 (2002). [CrossRef]
29. K. Yamashita, A. Kitanobou, M. Ito, E. Fukuzawa, and K. Oe, “Solid-state organic laser using self-written active waveguide with in-line Fabry-Pérot cavity,” Appl. Phys. Lett. 92(14), 143305 (2008). [CrossRef]
30. K. Yamashita, M. Ito, E. Fukuzawa, H. Okada, and K. Oe, “Device parameter analyses of solid-state organic laser made by self-written active waveguide technique,” J. Lightwave Technol. 27(20), 4570–4574 (2009). [CrossRef]
31. U. Brackmann, Lambdachrome® Laser Dyes (Lambda Physik, 1994).
32. K. Yamashita, E. Fukuzawa, H. Okada, and K. Oe, “Fiber-to-fiber optical gain of polymer-based amplifier with self-written active waveguide,” Jpn. J. Appl. Phys. 48(10), 102406 (2009). [CrossRef]