Open Access
Add to BibSonomyAbstract
We report the realization and characterization of an optofluidic microlaser based on a Fabry-Perot resonator fabricated by exploiting two direct writing fabrication techniques: the femtosecond laser micromachining and the inkjet printing technology. In this way a standard Fabry-Perot cavity has been integrated into an optofluidic chip. When using rhodamine 6G dissolved in ethanol at concentration of 5∙10−3 mol/l, laser emission was detected at a threshold energy density of 1.8 μJ/mm2 at least one order of magnitude lower than state-of-the-art optofluidic lasers. Linewidth below ~0.6 nm was measured under these conditions with a quality factor Q~103. These performances and robustness of the device makes it an excellent candidate for biosensing, security and environment monitoring applications.
© 2016 Optical Society of America
1. Introduction
The story of optofluidic lasers started more than a decade ago with a nice demonstration given by Helbo et al. who were able to show lasing action from a microchannel where Rh6G was slowly flowing, getting a linewidth of 5.7 nm with a threshold of few hundreds of µJ/mm2 [1]. After that time a lot of work has been done on this topic leading to different solutions to get single mode emission [2], low threshold [3], tunability [4]. Some of these configurations have been successfully exploited for biosensing [5]. The interest in developing such devices relies in the great potential of being used in lab-on-chip (LOC) technology because integrable coherent light sources are essential for those kind of applications, mainly in biomedical, security and environment monitoring field. The first issue to be addressed to make them attractive for large scale fabrication is avoiding a complex manufacturing composed of many different steps. This concept has been recently underlined by demonstrating the monolithic nanofabrication of an optofluidic ring resonator by femtosecond laser writing [6]. A reliable and robust ring microlaser has been realized achieving a threshold fluence of ≈15 µJ/mm2 and a quality factor Q ≈3.3∙104. This demonstration comes after more than ten years after the first attempt to realize a laser cavity using a similar fabrication technique [7]. In this case the fabricated optical cavity was realized by four 45° micromirrors vertically buried inside the glass substrate formed by hollow structures and working as total internal reflectors. A 80 µm diameter microchannel crossing the cavity provided the active medium flow (Rh6G dye in ethanol). Laser action was observed above an energy density threshold of ≈16.6 µJ/mm2 with a linewidth of about 5 nm. As a matter of fact the most simple laser cavity based on a Fabry-Perot resonator has not yet been realized exploiting this fabrication method so far due to the difficulty of getting suitable high reflectivity mirrors that are both perfectly aligned independently of the length of the cavity and not in contact with the active medium solution avoiding any degradation phenomenon.
Besides the first report [1] a few examples of Fabry-Perot cavity in optofluidic microlaser have been reported realizing reflecting mirrors by coating with metallic films either the edge of fibers used to collect the signal [8,9] or the side of the microfluidic channel [10,11]. In all these cases multi-steps soft lithography was used to realize the microfluidic structures, their perfect alignment is not easy especially for long cavity and usually the metallic mirror was put into contact with the dye solution. Most of these devices were mainly addressed to demonstrate multicolor emission and tunability rather than to improve threshold conditions and emission linewidth, that was in all cases larger than 3-4 nm. These last two laser peculiarities are the key requirements to be fulfilled in order to increase the measurement sensitivity in LOC platform systems.
In this paper we report the first demonstration of an optofluidic laser based on a Fabry-Perot optical cavity “fully embedded” in the glass substrate realized by coupling the ink-jet technology, necessary to get a fine control of surface coating using a highly reflective metallic ink, with the femtosecond laser processing. By using a solution of Rh6G as active medium we were able to get reliable laser emission above a threshold of 1.8 µJ/mm2 and a linewidth as low as ≈0.6 nm getting a quality factor Q ≈103.
2. Experimental methods
2.1 Design and fabrication
The devices realization takes advantage from an innovative combination of two direct writing fabrication techniques: the femtosecond laser micromachining and the inkjet printing technology.
Different geometries have been fabricated and investigated although the most basic configuration of the proposed optofluidic microlaser is shown in Fig. 1. It is composed by a central microfluidic channel where the active material flows and two square section empty “basins” – located close to the dye recirculation microchannel – where metallic mirrors will be directly printed in a second step, Fig. 1(b). Since the aim of these “basins” is to facilitate the mirror printing, its size and shape have been chosen in order to facilitate the ink wettability of the inner surface (avoiding phenomena of clogging before the annealing procedure) and to ensure a printed mirror diameter at least double with respect to the fiber core. Integrated broad band optical fiber, inserted in the partial reflector side, and two microtubes (inlet and outlet) complete the microcircuit design in order to out-couple the emitted light and connect the chip to the macroscopic external world. In order to insert the optical fiber (as well as the microtube), the structure includes two conical channels reaching the basins and the microfluidic channel. The special conical geometry of these channels ensures optimum fiber (and tubes) alignment with the internal part of the chip, already during the insertion.
Fig. 1 Sketch of the device. a) Fabry-Perot long cavity geometry; b) micro mirror fabrication procedure c) optical microscope image of the zoom section.
The microfluidic structure is completely buried inside a pure fused silica substrate (JGS1, FOCtek), obtained after ultrafast laser irradiation followed by chemical etching [12–15]. As well-known this technique is based on the multiphoton absorption and consequent modification of the physical-chemical properties of the substrate. In a particular irradiance regime, a self-organized internal structure (nanograting) is formed with a period of λ/2n (n integer) depending on the wavelength λ and polarization writing pulse. Moreover these assemblies are generated in the perpendicular direction with respect to the polarization of the laser beam. Since the femtosecond laser pulse writing affects a small volume of the material (only in the beam focus) without thermal effects, it is possible to create inside the substrate structures with arbitral 3D geometry.
When the written microchip is put in a chemical etching solution (typically HF), it digs the irradiated zone more rapidly respect to the non-irradiated ones creating the microfluidic circuit (the reasons of this selective action are mainly due to the chemical and physical modification the material achieved by the femtosecond laser treatment).
The used laser micromachining system is equipped with a regenerative amplified mode-locked femtosecond laser source based on Yb:KGW active medium (PHAROS - Light Conversion) whose pulses at the fundamental wavelength of 1030 nm are characterized by duration of 240 fs, repetition rate up to 1 MHz and pulse energy up to 0.2 mJ. In order to work in the “nanograting regime” [12] the second harmonic at λ = 515 nm and 500 kHz repetition rate has been used with average power set at 200 mW. The laser light is statically focused inside the substrate (1 mm thickness) through a microscope objective (50X, Mitutoyo, NA 0.42) and the 3D structure is achieved moving the sample, placed on an high precision three axis air-bearing translation stage (Aerotech, ABL 1000 series) with a resolution of 2 nm, at the speed of 1 mm/s.
Concerning the etching step the irradiated sample is placed in HF aqueous solution at 20% concentration at controlled temperature (35 °C) for few hours. The obtained small device (4 mm x 4 mm) represents the microfluidic chip as shown in Fig. 2(b).
Fig. 2 a) Optical micrograph of the top view of the microfluidic laser; b) photo showing the fabricated chip and the cavity with the laser dye.
Taking advantage from the optical fiber access, two broadband micro reflectors are fabricated in bulk structure by means of a controlled dispensing of silver organic complex inks (TEC-IJ-010 InkTec Co., Ltd.) using an inkjet printer (DIMATIX - Fujifilm). By adjusting the volume of the dispensed inks, one mirror was realized as completely reflective while the other one, to be used as output mirror, was realized has partial reflector, with a possible fine-tuning of the reflectivity in the range from 60% to 95%. In order to remove the solvents and make sintering of the silver film the sample is heated at 150 °C for 5 minutes on a hotplate. Finally the device must be connected to the external world, so two microfluidic tubes (external diameter 360 μm) are inserted into their specific compartments and fixed with UV-curable glue (NOA 65). Following the same procedure, an optical fiber is inserted and fixed in the semitransparent mirror arm in order to directly collect and out-couple the laser light. In this way a ready-to-use monolithic device has been obtained. In the case here reported the optical cavity length is of 1.6 mm, the average square channel section is of 200 μm and the output mirror reflectivity is about 92% at 570 nm. The actual device is shown in Figs. 2(a) and 2(b).
2.2 Optical characterization
In order to test the optofluidic laser performances, the glass chip was fixed on a x-y-z translation stage and properly plugged with inert tubes connected to a high performance syringe pump. It was providing a stable flow (3.3 μl/minute) of the solution made by RhG6 dissolved in ethanol at concentrations ranging from 1∙10−2 to 1∙10−4 mol/l. The used flow rate guarantees complete replacement of new dye molecules in the cavity volume between two succesive pumping pulses, in order to avoid any bleaching or quenching effect. The fiber optic coming from the glass chip was connected to an Ocean Optics spectrometer with resolution of 0.2 nm and the signal analyzed by a computer. The optofluidic laser was pumped by the second harmonic of a pulsed Nd-YAG laser at 532 nm and pulse duration ~5 ns. The pump beam power was controlled by means of neutral density filters and fine-tuned by rotating a half-wave plate placed before a polarizer providing vertical polarization of the beam. A beam splitter was used to monitor the intensity of each laser pulse by a fast photodetector. After the beam splitter, a cylindrical lens (focal length f = 5 cm) and a diaphragm allowed getting a focal spot suitable to overlap the optofluidic laser cavity. The reference photodetector was previously calibrated through measurements of the beam power by a power meter located in the position of the glass chip. In this way it was possible to monitor the energy of each pumping pulse. A careful measurement of the laser spot size was carried out by the knife-edge technique using a razor blade in front of a detector. The focused spot resulted to have a dimension of 2.2 × 0.09 mm2 ≈0.2 mm2. A sketch of the experimental set-up is given in Fig. 3.
3. Results and discussion
Measurements were performed at different pumping power and different dye concentrations. An example of the appearance of the laser action over the dye spontaneous emission band is given in Fig. 4.
Fig. 4 Typical emission spectrum increasing the pump power. Data correspond to dye concentration of 10−4 mol/l.
Actually the transition from the spontaneous to the laser emission can be easily observed in the less efficient cases, while when the laser threshold is very low, spontaneous emission cannot be easily observed since it is mixed with the detection noise. Figure 5 depicts the evolution of the spectrum linewidth showing the narrowing effect at the threshold of ~22 μJ/mm2. An optimal performances of this device were obtained with a careful alignment to get the best matching between the pump beam spot size and the dye microfluidic channel. Under these conditions the measured threshold values ranged from 1.8 μJ/mm2 to 22 μJ/mm2 depending on the dye concentration.
Fig. 5 Full width half maximum of the stimulated emission bandwidth vs pumping energy density. Data correspond to dye concentration of 10−4 mol/l.
Figure 6 displays the laser output vs the energy density of the pumping pulse for a Rh6G concentration of 5∙10−3 mol/l that showed the lowest threshold of 1.8 μJ/mm2. In Fig. 7 it is shown the typical emission spectrum above threshold for this concentration: such narrow linewidth (Δλ < 0.6 nm) slightly increases to values Δλ < 0.9 nm by increasing the pump power.
Fig. 6 The laser emission spectrum for the same device above threshold for dye concentration of 5∙10−3 mol/l.
Concerning the spectral behavior in some cases we have even observed a narrower linewidth of Δλ≈0.4 nm, but at the moment this result is not reliable as those reported in Figs. 4 and 7. In fact we should stress that these results have been obtained several times.
However, the measured threshold in all cases ranges from values lower than 2.0 μJ/mm2 to 25 μJ/mm2 with linewidth generally below 1 nm. The corresponding threshold energy of the pumping pulse ranges from 350 nJ to 5 μJ. Actual threshold is one order of magnitude or more lower than the typical one of the state-of-the-art optofluidic lasers regardless of the cavity design.
In one single case a lower threshold of 0.1 μJ/mm2 has been reported for an optofluidic ring resonator using aqueous quantum dots as active medium [16]. While concerning the linewidth is definitely much narrower than in the other mentioned Fabry-Perot based optofluidic lasers (0.5 nm against 3-5 nm). It leads to evaluate the quality factor Q in the range 0.5 ÷ 1∙103 depending on the dye concentration. This result points out the advantage of the femtosecond micromachining technology for the cavity fabrication. In fact the high value of the quality factor means low losses due to the high precision achieved in the parallelism of the mirrors surfaces without the need of further adjustement. The all-fiber laser recently reported [17] allows high repetition rate with emission performances similar to our device. However, the fabrication steps look quite complex compared to our system and it appears of difficult integration with LOC technology.
For all these reasons the present device, not only overcome the performances of any up-to-date Fabry-Perot based micro-lasers cavities, but it is also competitive with most of the optofluidic lasers realized in the last decade. The obtained very narrow linewidth as well as its robustness allows to foreseeing possible integration the device for high sensitivity optical sensor. Moreover the Fabry-Perot cavity can work in a wide range of optofluidic laser emission wavelengths to be easily chosen by selecting the appropriate dye circulating in the microfluidic channel. The threshold around 2.0 μJ/mm2 is very close to enter in the regime of inexpensive high power LED pumping.
The novelty is also the use of the ink-jet technology for realizing the inner mirrors, that are not in contact with the fluid active medium as in many optofluidic Fabry-Perot configurations, thus they are not subject to degradation due to chemical interaction with such a fluid.
The monolithic microfabrication combined with the huge flexibility shown by the two coupled technique allow adjusting the device performance for a wide range of applications which otherwise would require cumbersome integrations and alignment of each component (optical, microfluidic, ….) at every specific use.
4. Conclusions
We have successfully designed and fabricated an optofluidic laser Fabry-Perot resonator in fused-silica glass substrate. Such achievement has been possible by exploiting two direct writing fabrication techniques: the femtosecond laser micromachining and the inkjet printing technology. In this way a standard Fabry-Perot cavity has been integrated into an optofluidic chip.
The manufacturing procedure of this device satisfactorily addresses the main issue to make it attractive LOC technology: multi-step and complex procedures are replaced by an easy fabrication and robustness of the device.
The excellent results obtained in the performances of the optofluidic laser in term of emission linewidth and threshold confirm the recent claim that the femtosecond writing technology can be successfully used to create high quality optical cavities. In addition the use of high resolution inkjet printing technology allows realizing passive optical elements for optofluidic circuits providing a novel tool for their fabrication.
The flexibility is the strength of these coupled techniques, allowing testing different laser cavity geometries and size. Further developments of this analysis involve optimization of mirrors performances, optimum coupling investigation and use of different active media. Exploitation of this device in optical sensing is also planned.
Acknowledgments
The Authors wish to thank Maddalena Binda, Ph.D. for deep and thorough inkjet facilities training and the initial support with the mirror printing.
References and links
1. B. Helbo, A. Kristensen, and A. Menon, “A micro-cavity fluidic dye laser,” J. Micromech. Microeng. 13(2), 307–311 (2003). [CrossRef]
2. G. Aubry, Q. Kou, J. Soto-Velasco, C. Wang, S. Meance, J. J. He, and A.-M. Haghiri-Gosnet, “A multicolor microfluidic droplet dye laser with single mode emission,” Appl. Phys. Lett. 98(11), 111111 (2011). [CrossRef]
3. W. Song, A. E. Vasdekis, Z. Li, and D. Psaltis, “Low-order distributed feedback optofluidic dye laser with reduced threshold,” Appl. Phys. Lett. 94(5), 051117 (2009). [CrossRef]
4. M. Gersborg-Hansen and A. Kristensen, “Tunability of optofluidic distributed feedback dye lasers,” Opt. Express 15(1), 137–142 (2007). [CrossRef] [PubMed]
5. X. Fan and S.-H. Yun, “The potential of optofluidic biolasers,” Nat. Methods 11(2), 141–147 (2014). [CrossRef] [PubMed]
6. H. Chandrahalim, Q. Chen, A. A. Said, M. Dugan, and X. Fan, “Monolithic optofluidic ring resonator lasers created by femtosecond laser nanofabrication,” Lab Chip 15(10), 2335–2340 (2015). [CrossRef] [PubMed]
7. Y. Cheng, K. Sugioka, and K. Midorikawa, “Microfluidic laser embedded in glass by three-dimensional femtosecond laser microprocessing,” Opt. Lett. 29(17), 2007–2009 (2004). [CrossRef] [PubMed]
8. Q. Kou, I. Yesilyurt, and Y. Chen, “Collinear dual-color laser emission from a microfluidic dye laser,” Appl. Phys. Lett. 88(9), 091101 (2006). [CrossRef]
9. G. Aubry, S. Meance, A.-M. Haghiri-Gosnet, and Q. Kou, “Flow rate based control of wavelength emission in a multicolor microfluidic dye laser,” Microelectron. Eng. 87(5-8), 765–768 (2010). [CrossRef]
10. Y. Yang, A. Q. Liu, L. Lei, L. K. Chin, C. D. Ohl, Q. J. Wang, and H. S. Yoon, “A tunable 3D optofluidic waveguide dye laser via two centrifugal Dean flow streams,” Lab Chip 11(18), 3182–3187 (2011). [CrossRef] [PubMed]
11. D. V. Vezenov, B. T. Mayers, R. S. Conroy, G. M. Whitesides, P. T. Snee, Y. Chan, D. G. Nocera, and M. G. Bawendi, “A low-threshold, high-efficiency microfluidic waveguide laser,” J. Am. Chem. Soc. 127(25), 8952–8953 (2005). [CrossRef] [PubMed]
12. M. Beresna, M. Gecevičius, and P. G. Kazansky, “Ultrafast laser direct writing and nanostructuring in transparent materials,” Adv. Opt. Photonics 6(3), 293–339 (2014). [CrossRef]
13. R. Osellame, H. J. W. M. Hoekstra, G. Cerullo, and M. Pollnau, “Femtosecond laser microstructuring: an enabling tool for optofluidic lab-on-chips,” Laser Photonics Rev. 5(3), 442–463 (2011). [CrossRef]
14. R. Taylor, C. Hnatovsky, and E. Simova, “Applications of femtosecond laser induced self-organized planar nanocracks inside fused silica glass,” Laser Photonics Rev. 2(1-2), 26–46 (2008). [CrossRef]
15. S. LoTurco, R. Osellame, R. Ramponi, and K. C. Vishnubhatla, “Hybrid chemical etching of femtosecond laser irradiated structures for engineered microfluidic devices,” J. Micromech. Microeng. 23(8), 085002 (2013). [CrossRef]
16. A. Kiraz, Q. Chen, and X. Fan, “Optofluidic lasers with aqueous quantum dots,” ACS Photonics 2(6), 707–713 (2015). [CrossRef]
17. R. M. Gerosa, A. Sudirman, L. S. Menezes, W. Margulis, and C. J. S. de Matos, “All-fiber high repetition rate microfluidic dye laser,” Optica 2(2), 186 (2015). [CrossRef]
