We present a single mode, single polarization, distributed feedback polymer dye laser, based on a short high order Bragg grating defined in a dye doped polymer layer between two secondary polymer layers. The Bragg grating is defined solely with standard I-line UV lithography. In this device we obtain single mode operation in a multimode structure by means of mode loss differentiation without using sub-wavelength structures. The laser is fabricated using micro-fabrication technology, is pumped by a pulsed frequency doubled Nd:YAG laser, and emits light in the chip plane at 551.39 nm, with a FWHM linewidth below 150 pm.
© 2006 Optical Society of America
Low-cost, micro-fabricated lasers may find numerous applications in interference based sensor systems  and lab-on-a-chip microsystems [2, 3, 4]. Polymers are in general a preferred choice of material for such applications, as they are relatively cheap and easy to process, they may be synthesized with high chemical resistance and specific functionalities.
The first polymer based solid state dye lasers were demonstrated in the late 1960’s [5, 6, 7], and this type of devices has subsequently been widely studied as candidates for cost-efficient, maintenance-free tunable laser sources. A recent review on the further development of solid state polymeric dye lasers is given by Singh et al. . A variety of approaches for micro-fabricated polymer based dye lasers with integrated optical feedback have been demonstrated, see [9, 10, 11, 12, 13, 14, 15, 16, 17, 18] and references therein. Among the various approaches, thin film distributed feedback (DFB) techniques have proven particularly successful for obtaining single mode lasing [11, 17, 18]. Single mode DFB lasers are normally based on a low order Bragg reflection  and a π/2 phase-shift element , which in the visible range requires a spatial modulation on sub-micron scale [11, 17].
In a previous work , we have demonstrated single mode lasing operation in a DFB resonator, based on a short, high order Bragg grating. In this device , which was fabricated by standard (I-line) UV lithography in the negative tone resist SU-8  we obtained single-mode operation in a multimode structure by means of transverse-mode discrimination with an-tiguiding segments. The device , was an optically pumped, micro-fluidic dye laser, where the Rhodamine 6G laser dye, dissolved in ethylene-glycol was pumped through a microfluidic channel, passing the laser resonator on the chip. We have also used SU-8 as a polymer matrix for optically pumped, solid state dye lasers, using UV , electron beam  as well as X-Ray  lithography on SU-8 resist, doped with Rhodamine 6G laser dye. The refractive index of the Rhodamine 6G doped SU-8 is observed to increase with increasing dye concentration, which was used to realize single mode planar waveguides with a dye-doped SU-8 core sandwiched between pure SU-8 buffer and cladding layers .
In this paper, we present a solid state, single mode DFB polymer dye laser, fabricated by standard (I-line) UV lithography in dye doped SU-8, see Figs. 1(a) and 1(b). The laser is based on a short high order Bragg grating in Rhodamine 6G doped SU-8 on top of a pure SU-8 cladding layer. The laser structure can be placed on any flat substrate, suitable for spin-coating with SU-8. In the present device, we use a silicon wafer substrate.
A film of polymethylmethacrylate (PMMA) (nPMMA = 1.49) is spin-coated on top of the SU- 8 (refractive index n SU-8 ~ 1.6), in order to confine the light via planar waveguiding. Without the PMMA layer on top, in which case the SU-8 bars would be exposed to air, the feedback reflections would be too large, and the light would not be confined in between the dye doped SU-8 bars.
The three-layer polymer film forms a complex planar waveguide structure, with an embedded periodic reflection. The waveguide supports several propagating TE-TM modes, but different modes experience different loss traversing from one dye doped SU-8 segment to the next. This loss differentiation mechanism implies, that only the fundamental TE mode supports lasing. Furthermore, the structure gives rise to a modulation in intensity of different Bragg reflection orders, which increases the effective mode distance in the resonator, enabling single mode lasing in the microstructured device.
The photograph in Fig. 1(a) shows a typical polymer laser device. The laser structure on the silicon chip is 1 mm wide and 4.87 mm long. The layer structure making up the laser (Fig. 1(b)) consists of a 500 μm silicon substrate, a 7 μm pure SU-8 film, a 6 μm thick structure of dye doped SU-8 which is covered by a 4 μm layer of PMMA. The doped SU-8 is patterned into arrays of 150 bars, each 12.6 μm wide and 1 mm long, with a distance between the bars of 20 μm. The 150 bars of dye doped SU-8 make up a high order Bragg grating operating as an optical resonator. A π/4 phase shift in the middle of the structure ensures a single resonance for each Bragg reflection order .
Since the optical path length in one period of the Bragg grating is 49.96 μm, the spacing between the individual Bragg reflection orders (and thus resonances) will be 3.3 nm. However, since the phase evolution as function of wavelength is different in the doped SU-8 region and the PMMA region in a single grating period, the resonance modes caused by the Bragg reflection orders will be modulated in intensity. This modulation can be seen from Fig. 2 where the round trip loss as function of wavelength for a laser device is calculated. The calculation is made for the lasing threshold condition where one of the peaks has zero round trip loss. The modulation of the strength of the resonances increases the effective mode distance in the resonator, and enables lasing in a single Bragg order mode even though the gain profile of the Rhodamine 6G laser dye is quite broad (up to 50 nm FWHM ).
The dye doped SU-8 has a slightly higher refractive index than the undoped SU-8 buffer layer due to the presence of Rhodamine 6G dye. The refractive index difference lies between 1∙10-3 and 2∙10-3 at 633 nm according to , however the difference is likely to be somewhat higher closer to the resonance frequency of the dye molecules due to the phenomenon of anomalous dispersion (cf. with Ali et al.  for Rhodamine B). Since the refractive index of PMMA is much lower than for SU-8, the dye doped SU-8 forms the core of a planar waveguide.
The waveguide will be able to support a few propagating TE-TM modes, the number of which depends on the exact refractive index difference between the doped and the undoped SU-8. However, the loss for the different modes induced by the passage through the PMMA region of a grating period depends strongly on the TE-TM mode number. Therefore only one of the TE-TM modes is likely to partake in lasing action, ensuring single mode operation of the laser.
By modelling we have found the mode dependent loss for the case where the refractive index difference between doped and undoped SU-8 is 2∙10-3. With the assumed refractive index difference, the waveguide is able to support two TE-TM modes. Since the measurements revealed lasing only in TE modes, the following calculation ignores TM modes.
The calculation was performed with a finite difference beam propagation method (FDBPM from ), using the fundamental or the first TE mode as initial condition along with a complex refractive index matrix (1600 × 1600 elements) representing one period of the grating structure (see Fig. 3(a)). The shape of the PMMA covering the SU-8 was found from measurements. The squared norm of the field for the propagated modes is shown in Fig. 3(b) and Figure 3c.
The squared norm of the coupling coefficients, cmn, express the amount of energy in mode n being coupled into mode m after propagating one grating period. The coupling coefficients are calculated from the initial field, Un, propagated to Un, p, and coupled into the waveguide mode field distribution Um in the next waveguiding doped SU-8 region.
The squared norm of the coupling coefficients for the present parameters are given in Table 1, from where it is seen that the fundamental mode has the lowest loss caused by propagation through one grating period. The loss is quite high (almost half the energy in the mode is lost) and although it can be countered by the gain of the laser dye, the threshold for lasing is expected to be high in this case.
The devices were fabricated on a 4” silicon substrate wafer. The first SU-8 buffer layer was deposited in a spin process (60 seconds at 3000 rpm, resist: SU-8(10)), pre-exposure baked (90 °C for 2 minutes), flood UV exposed (15 seconds at 9 mW/cm2) and post-exposure baked (90 °C for 15 minutes). This yielded a pure SU-8 film with a thickness of 6 μm.
The Rhodamine 6G perchlorate doped SU-8 solution (2.5 μM Rhodamine 6G per gram of solid SU-8) was deposited in a second spin process (60 seconds at 1500 rpm), pre-exposure baked (90 °C for 2 minutes) and UV exposed through a chromium mask (15 seconds at 9 mW/cm2). After exposure the sample was post-exposure baked (90 °C for 15 minutes) and developed in PGMEA (development time: 2.5 minutes), leaving behind 4 μm high structures of dye doped SU-8. Since the buffer SU-8 layer was already UV exposed, the development did not affect this.
The layer of PMMA (13% in anisole, 950K) was deposited in a spin process (30 seconds at 1000 rpm) and baked (90 °C for 2 min) to evaporate the solvent of the PMMA. The wafer was finally diced into chips, for characterization.
The surface of the PMMA layer was measured with a profilometer to determine how the PMMA was deposited around the dye doped bars of SU-8. Figure 4 shows the measured profile of the PMMA (lower curve). The upper curve in the figure shows the de-convoluted profile, where the radius of curvature of the profilometer needle has been taken into account (5 μm). The de-convoluted profile was used in the modelling above to find the mode coupling coefficients.
The optical characterization was performed using optical pumping of the devices, while picking up the emitted light with a fiber attached to a CCD based spectrometer. The optical pumping was obtained with a frequency doubled Nd:YAG laser operating at 532 nm, delivering 5 ns pulses with a repetition rate of 10 Hz.
Figure 5 shows a typical spectrum from a laser device. The observed line width corresponds to the spectrometer resolution of 0.15 nm, so the laser line width is not resolved (pump energy density 250 μJ mm-2).
Figure 5 also shows the output energy (in arbitrary units) from the dye laser as function of pump pulse energy density. The output energy was obtained by integrating the spectra measured for different pump pulse energy densities. A kink in the curve signifying the laser threshold is observed at 220 μJ mm-2. This threshold is relatively high compared to e.g. [17, 18], which is expected from our modelling, that predicted a rather lossy structure, with |c 00|2 = 0.526.
Two lasers with similar design as described above, were cut from a common wafer and tested. The output spectra can be seen in Fig. 6. The high similarity between the spectra, indicates that the design is tolerant to process variations, such that the exact output wavelength does not jump due to small variations in structure height and other structural parameters.
We have demonstrated a device design enabling single mode lasing in the visible range from a micro structured solid polymer DFB dye laser. The devices are defined with minimum lithographic features of 10 μm, which is easily obtained with standard I-line UV lithography.
Conventionally wavelength selective elements for single mode lasers require sub-micron resolution of the lithography, obtained via methods such as interference, electron beam or imprint lithography. In contrast, we have used I-line UV lithography and obtained the same kind of reproducible results with respect to single mode operation of the lasers.
The work was supported by the Danish Technical Research Council. (STVF, grant no: 26–02–0064).
References and links
01. L. Lading, L. B. Nielsen, and T. Sevel, ”Comparing Biosensors,” Proceedings of the IEEE Sensors 2002, 229–232 (2002) [CrossRef]
02. E. Verpoorte and N. F. De Rooij, ”Microfluidics meets MEMS,” Proceedings of the IEEE 91930–953 (2003) [CrossRef]
03. E. Verpoorte E, ”Chip visionoptics for microchips,” Lab on a Chip 342N52N (2003)
04. S. Balslev, A. M. Jorgensen, B. Bilenberg, K. B. Mogensen, D. Snakenborg, O. Geschke, J.P. Kutter, and A. Kristensen ”Lab-on-a-chip with integrated optical transducers,” Lab on a Chip, accepted for publication (2005)
05. B. H. Soffer and B. B. McFarlandContinuously tunable, narrow-band organic dye lasers, Appl. Phys. Lett. 10, 266267 (1967) [CrossRef]
06. O. G. Peterson and B. B. Snavely, Stimulated emission from flashlamp-excited organic dyes in polymethyl methacrylate, Appl. Phys. Lett. 12, 238240 (1967)
07. H. Kogelnik and C. Shank, Stimulated emission in a periodic structure, Appl. Phys. Lett. 18, 152154 (1971)
08. S. Singh, V. R. Kanetkar, G. Sridhar, V. Muthuswamy, and K. Raja, Solid-state polymeric dye lasers, J. Luminescence 101, 285291 (2003) [CrossRef]
09. C. Hu and S. Kim, ”Thin-film dye laser with etched cavity,” Appl. Phys. Lett. 29, 582–585 (1976) [CrossRef]
10. Y. Li, M. Sasaki, and K. Hane, ”Fabrication and testing of solid polymer dye microcavity lasers based on PMMA micromolding,” J. Micromech. Microeng. 11, 234–238, (2001)
11. Y. Oki, T. Yoshiura, Y. Chisaki, and M. Maeda, ”Lasers and Laser Optics - Fabrication of a distributed-feedback dye laser with a grating structure in its plastic waveguide,” Appl. Opt. 41, 5030–5035,(2002) [CrossRef] [PubMed]
12. D. Nilsson, T. Nielsen, and A. Kristensen, ”Molded plastic micro-cavity lasers,” Microelectron. Eng. 73–74, 372–376 (2004) [CrossRef]
13. D. Nilsson, T. Nielsen, and A. Kristensen, ”Solid State Micro-cavity Dye Lasers Fabricated by Nanoimprint Lithography,” Rev. Sci. Instrum. 75, 4481–4486 (2004) [CrossRef]
15. M. Gersborg-Hansen, S. Balslev, N. A. Mortensen, and A. Kristensen ”A coupled cavity micro fluidic dye ring laser,” Microelectron. Eng. 78–79, 185–189 (2005) [CrossRef]
16. J.C. Galas, J. Torres, M. Belotti, Q. Kou, and Y. Chen, ”Microfluidic tunable dye laser with integrated mixer and ring resonator,” Appl. Phys. Lett. 86, 1–3 (2005) [CrossRef]
18. S. Balslev, T. Rasmussen, P. Shi, and A. Kristensen, ”Single mode solid state distributed feedback dye laser fabricated by grey scale electron beam lithography on dye doped SU-8 resist,” J. Micromechanics Microeng. 15, 2456–2460 (2005) [CrossRef]
19. R. G. Hunsperger, ”Integrated Optics: Theory and Technology,” Fifth edition, Springer series in optical sciences, Springer-Verlag, Heidelberg (2002)
20. S. L. McCall and P. M. Platzman, ”An optimized p/2 distributed feedback laser,” IEEE J. Quantum Electron. 21, 1899–1904 (1985) [CrossRef]
21. K. Y. Lee, N. LaBianca, S. Zolgharnain, S. A. Rishton, J. D. Gelorme, J. M. Shaw, and T. H. P. Chang, ”Micro-machining applications of a high resolution ultrathick photoresist,” J. Vac. Sci Technol. B 133012–3016 (1995) [CrossRef]
22. S. Balslev and F. Romanato, ”Functionalized SU-8 patterned with X-ray Lithography,” J. Vacuum Sci. Technol. B accepted for publication (2005) [CrossRef]
23. B. B. Snavely, ”Flashlamp-excited organic dye lasers,” Proc. IEEE 57, 1374–1390 (1969) [CrossRef]
24. M. A. Ali, J. Moghaddase, and S. A. Ahmed, ”Optical properties of cooled rhodamine B in ethanol,” J. Opt. Soc. Am. B 8, 1807–1810 (1991) [CrossRef]
25. H. J. W. M. Hoekstra, G. J. M. Krijnen, and P. V. Lambeck, ”Efficient Interface Conditions for the Finite Difference Beam Propagation Method,” J. Lightwave Technol. 10, 1352–1355 (1992) [CrossRef]