We demonstrate an opto-fluidic ring resonator dye laser using highly efficient energy transfer. The active lasing material consists of a donor and acceptor mixture and flows in a fused silica capillary whose circular cross section forms a ring resonator and supports the whispering gallery modes (WGMs) of high Q-factors (>107). The excited states are created in the donor and transferred to the acceptor through the fluorescence resonant energy transfer (FRET), whose emission is coupled into the WGM. Due to the high energy transfer efficiency and high Q-factors, the acceptor exhibits a lasing threshold as low as 0.3 μJ/mm2. We further analyze the energy transfer mechanisms and find that non-radiative Förster transfer is the dominant effect to support the acceptor lasing. FRET lasers using cascade energy transfer and using quantum dots (QDs) as the donor are also presented. Our study will not only lead to development of novel microfluidic lasers with low lasing thresholds and excitation/emission flexibility, but also open an avenue for future laser intra-cavity bio/chemical sensing.
© 2007 Optical Society of America
In the past few years, microfluidic dye lasers have been a very active area in opto-fluidics for lab-on-a-chip devices and compact laser sources [1,2]. Various optical cavities have been explored, including Fabry-Pérot cavities [3,4], distributed feedback structures [5–8], and ring resonators [9–14]. Recently, we have developed a unique liquid core optical ring resonator (LCORR) based on a thin-walled fused silica capillary that inherently integrates the ring resonators with the microfluidics (see Fig. 1(A)) [15,16]. The LCORR has extremely high Q-factors (>107), thus providing an excellent feedback for low-threshold lasing. A lasing threshold of 1 μJ/mm2 has been achieved with R6G and the laser emission can be out-coupled evanescently by a waveguide in touch with the LCORR for convenient light guiding .
In those microfluidic lasers, the gain media, i.e., dyes, are usually directly excited. As a result, the selection of the dye is significantly limited by the availability of pump lasers. Furthermore, due to small size of the microfluidic channel, the absorption of pump light is weak. Although the low absorption efficiency may be alleviated by using a higher dye concentration, the existence of dye self-absorption leads to an increased lasing threshold, severely compromising the laser performance.
These issues can be overcome with fluorescence resonant energy transfer (FRET), in which the acceptor laser is pumped by the energy transferred from the donor. Among all FRETs, non-radiative Förster transfer is the most important mechanism [17,18]. During the Förster transfer, the energy in an excited donor is transferred directly to an acceptor in the close proximity without mediation of any photon. The Förster transfer efficiency between a single donor and a single acceptor, EF, is highly dependent on the donor-acceptor distance: EF = R 0 6/(R 0 6 + r 6), where r is the donor/acceptor separation. R 0 is the Förster distance [17,18]. For most dye-dye pairs, R 0 ranges from 2–10 nm , corresponding to a solution concentration in the range of μM-mM, which is the typical concentration for microfluidic dye lasers [8,12,13,16]. Therefore, efficient energy transfer can be employed to achieve lasing in the acceptor. In addition, since the absorption band and emission band in a FRET laser are much farther apart than in a conventional single dye laser, the donor absorption at the acceptor lasing wavelength can be virtually ignored, resulting in a lower lasing threshold in the acceptor. In fact, the Förster energy transfer has been extensively used in solution- and polymer-based dye lasers to tune the laser emission into the near infrared region, to enhance the pump efficiency while avoiding self-absorption problems, and to lower the lasing threshold [19–22].
In this paper, we demonstrate the highly efficient FRET microfluidic dye laser based on the LCORR. The concept of this LCORR FRET laser system is illustrated in Fig. 1(A). The donor and acceptor are co-flowed through the LCORR and the energy from the donor is transferred to the acceptor whose emission is coupled to the WGMs of high Q-factors. We first perform detailed investigation on the LCORR FRET laser using a dye-dye pair as a model system. It is found that a very low acceptor lasing threshold (0.3 μJ/mm2) can be achieved due to the high energy transfer efficiency between the donor and acceptor, and to the high Q-factor of the LCORR. The energy transfer mechanisms are analyzed, showing that the non-radiative Förster transfer dominates in supporting the acceptor lasing. Lasers that use cascade FRET involving three dyes and that use quantum dots (QDs) as the donor are also explored. Finally, one of the important applications of the LCORR FRET laser, intra-cavity sensing, is discussed.
2. Experiment and results
The LCORR fabrication and the experimental setup have been described earlier in Ref. . Briefly, the LCORR with an outer diameter of 75 μm and a wall thickness of 5 μm is made by rapidly stretching a fused silica preform using a computer-controlled capillary pulling station (Fig. 1(B)) [15,16]. A subsequent slight HF etching will further reduce the wall thickness down to 4 μm . The Q-factor is characterized by a tunable diode laser and Q = 107 is obtained at 690 nm .
The dye mixture in methanol is passed through the LCORR at a flow rate of 10 μL/min. For LCORR FRET lasers, the donor is excited by a 10-ns pulsed laser (Opolette, 20 Hz repetition rate) at 526 nm. The laser is loosely focused through a cylindrical lens so that an approximately 1 mm wide band along the LCORR is illuminated. The dye emission is collected in free space and is detected by an Ocean Optics spectrometer (USB4000, 3.7 nm spectral resolution) positioned 90-degrees off the incident beam.
We first use R6G and LDS 722 as the donor and acceptor. Figure 2 shows the characterization of their FRET behavior in free space, which is crucial in understanding and development of the FRET lasers. For this characterization, the donor and acceptor in methanol are placed in a cuvette and the donor is excited with a low power CW laser at 532 nm. As shown in Fig. 2(A), in the absence of R6G, no LDS 722 emission is observed. When 0.1 mM of R6G is added, the acceptor emission emerges with the peak at 700 nm, and increases with the increased acceptor concentration. The energy transfer efficiency for each acceptor-to-donor ratio (A/D) is plotted in Fig. 2(B), which can be derived by: η = 1-ID/ID0, where ID and ID0 are the donor intensity in the presence and absence of the acceptor, respectively. According to Ref. , η is related to the Förster distance, R0, by:
where c is the acceptor concentration. c0 is the critical concentration that corresponds to one acceptor molecule in a sphere of the radius of R0. Curve fit in Fig. 2(B) yields c0 = 1.7 mM, which results in a Förster distance of 6.2 nm . 50% transfer efficiency is obtained when the acceptor concentration is 0.83 mM (i.e., A/D = 8.3), reflecting a highly efficient energy transfer between the donor and acceptor. The quantum yield of the acceptor, ϕ, can also be estimated by:
where IA is the acceptor intensity. For LDS 722, ϕ = 7%, much lower than that for R6G, which has nearly unity quantum yield.
The LCORR FRET laser using R6G and LDS 722 is demonstrated in Fig. 3(A). First, 0.1 mM pure R6G solution is passed through the LCORR. The R6G lasing emission is observed at 562 nm, which is typical for R6G at this concentration . When the acceptor (LDS 722) is added (while the pump intensity remains the same), the donor (R6G) lasing decreases. Meanwhile, prominent peaks emerge at the longer wavelength side of the acceptor emission spectrum (715, 724, and 733 nm), indicating that the lasing threshold is first reached at those wavelengths. Each of these peaks consists of laser emission from different WGMs, as indicated by the modulation on the peak intensity. However, those finer peaks can not be fully resolved due to the low spectrometer resolution. When A/D further increases, the donor lasing is completely quenched and acceptor lasing emission increases. Additional peaks at shorter and longer wavelengths appear, suggesting that the lasing gain is achieved at those wavelengths. Note that the lasing peaks result from the donors and acceptors that lie within 100 nm from the LCORR interior surface, as only the emission from those molecules can be coupled into the WGM whose evanescent field has a penetration depth of approximately 100 nm into the core. Therefore, the effective volume of sample that contributes to the LCORR FRET laser is only 20 pL, 0.6% of the total illuminated volume (~ 3.3 nL).
Figure 3(B) shows the lasing threshold for the peak at 724 nm for two acceptor concentrations (1.5 mM and 2.0 mM) when the donor concentration is kept at 0.1 mM. For a comparison, the intensity of R6G at 562 nm is also acquired under the same experimental condition in the absence of LDS 722. The lasing threshold is approximately 0.3 μJ/mm2 and 0.5 μJ/mm2 for 2 mM and 1.5 mM LDS 722, respectively, which is over ten times lower than that of R6G at 0.1 mM concentration and is even two to three times lower than the lasing threshold for R6G of the similar concentration (~ 1 μJ/mm2). Such a low threshold can only be achieved through the indirect yet highly efficient energy transfer. Given the low quantum yield of LDS 722 (7% vs. 100% in R6G), a much higher lasing threshold would be expected if LDS 722 were directly excited.
Now we investigate the energy transfer mechanism involved in the LCORR FRET laser. In addition to the short-ranged non-radiative Förster transfer, the energy transfer between donor and acceptor can be achieved through the mediation of the WGM [23,24]. In this cavity-assisted radiative energy transfer, the donor emission is first coupled into the WGM that stores the photons for extended time for subsequent absorption by acceptors. For a ring resonator of high Q-factor, highly efficient radiative energy transfer has been observed . To determine which effect is dominant, we carry out a few control experiments. The first one is shown in Fig. 4(A) where the A/D remains constant and the pump power is set to be below, near, and well above the R6G (donor) lasing threshold. In all three cases, the acceptor lasing is achieved. Here we use γ = (1+ID/IA)−1 to characterize the effective photon conversion efficiency from the donor to the acceptor, as η may not be the best measure to describe the energy transfer in a non-linear lasing system. When the pump is below the lasing threshold, only broad band fluorescence is observed while LDS 722 still lases. As discussed earlier, we are interested only in the dye molecules lying within 100 nm of the LCORR interior surface. Therefore, for those donors and acceptors, γ is nearly 100%, meaning that all the donor excited states are converted to the lasing photons in the acceptors. When the pump power increases, the R6G starts to lase, as indicated by the lasing peak around 560 nm above the broad fluorescence background. However, the acceptor lasing intensity remains nearly the same, making γ drop to 91%. When the pump is well above the R6G threshold, a much stronger lasing peak occurs. Acceptor lasing intensity increases, but at a much lower percentage, and γ becomes only 63%. This suggests that the R6G lasing process competes with the energy transfer process. When R6G lases, the radiative decay of R6G becomes faster than in fluorescence because of stimulated emission, resulting in less energy transfer to the acceptor. This result agrees with the non-radiative Förster transfer model, but contradicts the radiative cavity-assisted energy transfer model, in which the acceptor intensity would be expected to increase proportionately when the donor lases.
Additional evidence to confirm the dominant effect of the non-radiative Förster transfer is plotted in Fig. 4(B), in which the R6G concentration changes while the pump intensity and the acceptor concentration remain constant. R6G lasing is achieved in all three R6G concentrations. However, γ decreases with the increased R6G concentration (hence, a lower A/D), despite the increase in the R6G intensity. This once again supports the non-radiative Förster transfer model and contradicts the cavity-assisted energy transfer model.
To further demonstrate the LCORR FRET laser capability, we show in Fig. 5(A) the lasing achieved with cascade FRET, where the energy is transferred from Coumarin 480 (donor) to R6G, and then to LDS 722. Strong acceptor laser emission emerges when the ratio between LDS 722 and Coumarin is 10:1. R6G emission is nearly invisible, indicative of a highly efficient energy transfer between R6G and LDS 722. As a control, no LDS 722 emission is observed in the absence of R6G, showing that Coumarin and LDS are completely decoupled. Therefore, a high concentration of Coumarin can be used to enhance the pump light absorption efficiency without concern about the donor absorption at the acceptor lasing wavelength.
Although the absorption band and emission spectral span of the microfluidic laser is significantly extended using FRET between dye pairs, the absorption of the donor dye still requires a particular light source that fits its absorption band. Recently, semiconductor QDs have been successfully employed as the donor [25–29]. QDs have much better chemical/photo stability, higher quantum yield, and higher absorption cross section than dyes. Moreover, all sizes of QDs can be excited by a single blue light source and their fluorescence can be tuned by changing the QD size to match the absorption band of various acceptors [26,30]. Therefore, using QDs as a donor enables the maximal pump flexibility and multiple-emission capability in microfluidic lasers. Figure 6 presents the LCORR FRET laser with the core/shell structured QDs as the donor. Over 200 nm spectral separation between the pump and the laser emission is achieved without an intermediate dye. Due to the large QD absorption, only 200 nM QD is needed to achieve the sufficient energy transfer, which corresponds to only 4 atto-mole QDs.
In addition to a flexible, low threshold, and compact microfluidic laser, the LCORR FRET laser can also be used for ultra-sensitive intra-cavity bio/chemical sensing [31,32]. As compared to conventional FRET, the lasing action is a non-linear effect. As shown in Fig. 3(A), in the presence of the acceptor, the gain in the donor is reduced, resulting in an exponential decrease in the donor emission. Meanwhile, when the gain in the acceptor is achieved through the energy transfer from the donor, stimulated emission takes place in the acceptor, producing significantly higher intensity than spontaneous emission. Therefore, the conventional Förster transfer is tremendously enhanced in this lasing system. Generally, in FRET bio/chemical sensing, γ or IA/ID is used as a sensing signal, as it is a ratiometric measurement that significantly reduces the impact of the power variation in excitation sources. Figure 7 compares γ obtained from Fig. 2(A) and Fig. 3(A), showing that γ can be enhanced over 20 times at a low acceptor concentration for the donors and acceptors that participate in the LCORR FRET laser.
In practice, there exists a broad donor emission background resulting from the donor molecules in the central part of the LCORR, as shown in Fig. 3(A), which may interfere with the FRET measurement and reduce the benefit from LCORR laser enhanced FRET. However, this background can be completely removed by coupling the FRET signal out through an optical taper or waveguide in touch with the LCORR . As a result, only the light coupled into the WGM will be detected, thus allowing us to take full advantage of the enhanced FRET associated with the non-linear lasing action.
We have demonstrated a low threshold opto-fluidic ring resonator laser using highly efficient FRET, which can be achieved in the format of a two-dye pair, a multiple-dye cascade pair, and a QD-dye pair. We further find that the non-radiative Förster transfer plays a dominant role in the energy transfer. Our results will lead to development of novel microfluidic lasers and open a door to highly sensitive intra-cavity bio/chemical sensing.
The authors acknowledge the support from the 3M Non-Tenured Faculty Award and the Wallace H. Coulter Early Career Award. The authors also thank Dr. Ian M. White for discussion. Dr. S. Lacey is a visiting professor from Franklin & Marshall College.
References and links
2. C. Monat, P. Domachuk, and B. J. Eggleton, “Integrated optofluidics: A new river of light,” Nat. Photonics 1, 106–114 (2007). [CrossRef]
3. Q. Kou, I. Yesilyurt, and Y. Chen, “Collinear dual-color laser emission from a microfluidic dye laser,” Appl. Phys. Lett. 88, 091101 (2006). [CrossRef]
4. B. Helbo, A. Kristensen, and A. Menon, “A micro-cavity fluidic dye laser,” J. Micromech. Microeng. 13, 307–311 (2003). [CrossRef]
8. Z. Li and D. Psaltis, “Optofluidic Distributed Feedback Dye Lasers,” J. Sel. Top. Quantum Electron. 13, 185–193 (2007). [CrossRef]
9. H.-M. Tzeng, K. F. Wall, M. B. Long, and R. K. Chang, “Laser emission from individual droplets at wavelengths corresponding to morphology-dependent resonances,” Opt. Lett. 9, 499–501 (1984). [CrossRef] [PubMed]
11. A. Sennaroglu, A. Kiraz, M. A. Dündar, A. Kurt, and A. L. Demirel, “Raman lasing near 630 nm from stationary glycerol-water microdroplets on a superhydrophobic surface,” Opt. Lett. 32, 2197–2199 (2007). [CrossRef] [PubMed]
14. 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, 264101 (2005). [CrossRef]
16. S. I. Shopova, H. Zhou, X. Fan, and P. Zhang, “Optofluidic ring resonator based dye laser,” Appl. Phys. Lett. 90, 221101 (2007). [CrossRef]
17. J. R. Lakowicz, Principles of Fluorescence Spectroscopy, (Kluwer Academic/Plenum Publishers, New York City, New York,1999).
18. T. Förster, “Transfer mechanisms of electronic excitation,” Disc. Faraday Soc. 27, 7–17 (1959).
19. C. E. Moeller, C. M. Verber, and A. H. Adelman, “Laser pumping by excitation transfer in dye mixtures,” Appl. Phys. Lett. 18, 278–280 (1971). [CrossRef]
20. M. Berggren, A. Dodabalapur, R. E. Slusher, and Z. Bao, “Light amplification in organic thin films using cascade energy transfer,” Nature 389, 466–469 (1997). [CrossRef]
21. M. Berggren, A. Dodabalapur, and R. E. Slusher, “Stimulated emission and lasing in dye-doped organic thin films with Förster transfer,” Appl. Phys. Lett. 71, 2230–2232 (1997). [CrossRef]
22. M. I. Savadatti, S. R. Inamdar, N. N. Math, and A. D. Mulla, “Energy-transfer dye lasers,” J. Chem. Soc. Faraday Trans. 82, 2417–2422 (1986). [CrossRef]
24. S. Gotzinger, L. D. S. Menezes, A. Mazzei, S. Kuhn, V. Sandoghdar, and O. Benson, “Controlled Photon Transfer between Two Individual Nanoemitters via Shared High- Q Modes of a Microsphere Resonator,” Nano Lett. 6, 1151–1154 (2006). [CrossRef] [PubMed]
25. I. L. Medintz, A. R. Clapp, F. M. Brunel, T. Tiefenbrunn, H. T. Uyeda, E. L. Chang, J. R. Deschamps, P. E. Dawson, and H. Mattoussi, “Proteolytic activity monitored by fluorescence resonance energy transfer through quantum-dot-peptide conjugates,” Nat. Mater. 5, 581–589 (2006). [CrossRef] [PubMed]
26. A. R. Clapp, I. L. Medintz, and H. Mattoussi, “Förster Resonance Energy Transfer Investigations Using Quantum-Dot Fluorophores,” ChemPhysChem 7, 47–57 (2006). [CrossRef]
29. R. Gill, I. Willner, I. Shweky, and U. Banin, “Fluorescence Resonance Energy Transfer in CdSe/ZnS-DNA Conjugates: Probing Hybridization and DNA Cleavage,” J. Phys. Chem. B 109, 23715–23719 (2005). [CrossRef] [PubMed]
30. C. A. Leatherdale, W.-K. Woo, F. V. Mikulec, and M. G. Bawendi, “On the Absorption Cross Section of CdSe Nanocrystal Quantum Dots,” J. Phys. Chem. B 106, 7619–7622 (2002). [CrossRef]
32. A. W. Wun, P. T. Snee, Y. Chan, M. G. Bawendi, and D. G. Nocera, “Non-linear transduction strategies for chemo/biosensing on small length scales,” J. Mater. Chem. 15, 2697–2706 (2005). [CrossRef]