A fluorescent microdroplet was formed in elastomer to facilitate handling and wavelength tuning. A methanol solution of rhodamine 6G was put into a solidifying polysiloxane resin with a needle. Addition of surfactant was effective to stabilize the droplet. Being excited by a laser pulse (532 nm, >50 μJ/mm2, 5 ns), the droplet exhibited a whispering gallery mode emission in the 570–610 nm wavelength range. The resonance peaks shifted as the droplet diameter expanded by elastomer deformation.
©2008 Optical Society of America
Whispering gallery mode (WGM) emission from dye-doped spheres is useful for constructing low-threshold lasers and microsensors [1,2]. Organic dyes are excellent fluorescence emitters, since they exhibit high absorbance, high quantum efficiency, and a wide tuning range. Emission wavelength can be tuned by changing resonance parameters, e.g., sphere size, refractive index, and coupling efficiency with a surrounding material [3,4]. Wavelength tuning becomes easier, if a droplet is used in place of a solid sphere. Some researchers observed WGM resonance in droplets that were suspended in air or oil [3,5,6]. Tuning of WGM resonance was also conducted with droplets that were sprayed on a glass plate . When constructing optical devices, however, droplets have to be fixed in solid matrix to attain functional stability and handling facility. Recently, elastomers or gels have been used in optical devices to realize mechanical or electrical control of a periodic structure [8–11]. If a dye droplet is suspended in such deformable matrix, a tunable microlaser will be realized. In this work, we formed a droplet in elastomer, and observed fluorescence spectrum change during a deformation process.
2. Sample preparation and experiments
A fluorescent droplet was fabricated with a methanol solution of rhodamine 6G (0.5 mol/l) by using the process shown in Fig 1(a) Raw liquid of polysiloxane (Shin-Etsu Chemical, KE-103) was mixed with a curing agent, and the mixture was put into a plastic mold (inner area: 30×30 mm2) to 7 mm depth. This liquid solidified in 8 h and became a transparent elastomer. (b) On the way of solidification, i.e., 2 h after addition of the curing agent, a needle with the dye solution on its tip was inserted into the solidifying liquid. (c) When the needle was taken out of the liquid slowly, the dye solution remained in the liquid. The solution formed a sphere in self-organized manner, since hydrophilic methanol did not mix with hydrophobic polysiloxane. A droplet of 10-4 mm3 (100 pl), for example, created a sphere of ∼60 μm diameter. The droplet remained in the middle of the solidifying liquid, since the liquid was viscous enough at this stage. After solidification was complete, the elastomer was taken out of the mold.
The droplet in the elastomer exhibited a bright orange color for several hours after fabrication. As time passed, however, the color became darker, and finally, the droplet stopped fluorescence emission. We assumed two possible reasons for this phenomenon; i.e., some chemical reaction took place between rhodamine and polysiloxane, or methanol diffused into polysiloxane matrix leaving a condensed rhodamine particle. In either case, isolating the droplet from the matrix was thought to prevent the fluorescence deterioration. We therefore added a surfactant (Merck, tween 20) to the dye solution before putting it into the solidifying liquid. The surfactant was assumed to form a protection layer (micelle) on the droplet surface, with the hydrophilic group inside and the hydrophobic group outside. Actually, the droplet remained stable in the elastomer without changing its color, when the surfactant concentration was 0.5 mol/l. The surfactant with lower concentration was not effective to prevent the droplet deterioration.
Figure 2(a) shows a micrograph of a droplet fabricated in this manner. Droplet diameter d was controlled between 50 and 140 μm by changing the dipping depth of the needle tip in the dye solution. When the elastomer was pressed from the sides, the droplet deformed into an ellipsoid, as shown in Fig. 2(b). We irradiated a pump laser beam to the droplet from the top, as shown in Fig. 2(c), and observed fluorescence from the side by using a multichannel spectrometer (Ocean Optics, HR2000). The pump beam was a frequency-doubled Nd:YAG laser (Continuum, Surelite II, wavelength: 532 nm, pulse duration: 5 ns, beam diameter: 5 mm). The laser beam was irradiated to the droplet without focusing. The beam fluence was controlled between 50 and 500 μJ/mm2 by using an attenuator.
3. Experimental results
The methanol solution of rhodamine 6G usually exhibited a broad fluorescence spectrum, as exemplified by the gray line in Fig. 3. As the black lines show, however, a cluster of narrow peaks appeared when a droplet was excited by a laser pulse of >50 μJ/mm2. A droplet of ∼60 μm diameter exhibited periodic peaks in the vicinity of 575 nm. This spectrum was similar to those of the WGM resonance reported in previous papers [2–7]. As the droplet diameter became larger, the resonance peaks became higher and shifted to longer wavelengths, i.e., ∼585 nm (80 μm) or ∼600 nm (90 μm).
Next, a sample was pressed from the side with a micrometer, as shown in Fig. 2(c); i.e., an elastomer of 30 mm width was deformed by Δw=0.03 or 0.05 mm. Figures 4(a) and 4(b) show fluorescence spectra that were measured during the deformation process. The resonance peaks of a 60 μm droplet shifted to longer wavelengths by ∼0.4 nm as the elastomer deformed by 0.03 mm. Similarly, the peaks of an 80 μm droplet shifted by ∼0.2 nm due to 0.05 mm deformation. Figures 4(c) and 4(d) show relations between peak wavelengths and elastomer deformation (Δw). The circles and bars show averages and deviations of measured data taken by repeating measurements five times for each deformation. All peaks shifted to longer wavelengths as the elastomer deformed. As shown in Fig. 2, elastomer deformation caused a droplet to expand in the plane perpendicular to the deformation direction. Therefore, the optical path of the WGM extended in this plane, thereby causing the resonance peaks to shift to longer wavelengths.
In a droplet of diameter d, WGM resonance peaks appear at wavelengths
where n denotes the refractive index of the droplet (n=1.4). Therefore, the droplet diameter can be evaluated from wavelengths of two adjacent peaks, λm and λm+1, as
When we deformed the elastomer further to Δw=1 mm, we could not identify the corresponding peaks before and after deformation, since the peak shift was larger than the peak spacing (free spectral range). As Fig. 5(a) shows, however, we observed notable decrease in peak spacing; i.e., the peak spacing λm−λm+1 decreased as the droplet diameter increased [Eq. (2)]. Figure 5(b) shows droplet diameters that were evaluated from the measured peak wavelengths. As the elastomer deformed by Δw=1 mm, a droplet of ∼90 μm diameter expanded by ∼4 μm (∼4%) in the plane perpendicular to the elastomer deformation. Similarly, a droplet of ∼140 μm diameter expanded by ∼6 μm (∼4%). These results suggest that the resonance peak wavelengths also increase by ∼4%; i.e., a peak at 600 nm possibly shifts to ∼624 nm. However, such a large peak shift is difficult to observe, since resonance of higher-order modes becomes dominant due to wavelength dependence of the optical gain and loss in a droplet.
WGM emission usually takes place when the refractive index of a sphere is higher than that of a surrounding material, i.e., when the total reflection takes place at the sphere surface. As regards the current experiment, however, the refractive index of the methanol solution (n=1.4) is lower than that of polysiloxane (n=1.5). If the incident angle is ∼89°, for example, this index difference causes the Fresnel reflection of 80–90% at the droplet surface, which seems to be too low to excite WGM resonance. It has been reported in the recent paper that a high-refractive-index surface layer concentrates the WGM field in the vicinity of the droplet surface . We assume that the surfactant layer (n>1.5) is effective to confine the WGM radiation in the droplet. The surfactant layer thickness, however, is difficult to evaluate, since some amount of surfactant seems to disperse in the polysiloxane matrix during the needle insertion process. Additional experiments are needed to study how the surfactant layer affects WGM resonance.
In the current experiment, elastomers were pressed in the direction perpendicular to the excitation and observation directions, lest the mechanical stage should interfere the light beams. Elastomers, of course, can be pressed in other directions, if a press apparatus is so designed as to transmit the light beams. Further, one will be able to deform the elastomer by using an electrostatic force, if electrodes are deposited on the elastomer surfaces .
As Figs. 4 and 5 show, the resonance peaks tended to become lower, as the deformation Δw became larger. In addition, we occasionally observed the resonance peaks to disappear during the deformation process. These phenomena were probably caused by asymmetry or non-uniformity of deformed droplets.
A droplet formation with a needle provides a simple fabrication process of a deformable microcavity. This method, however, does not allow accurate control of the droplet size; i.e., the liquid volume on the needle tip is difficult to adjust as desired by only changing the needle dipping depth in the dye solution. Also, a droplet smaller than 50 μm diameter is difficult to form by this method. Recent progress in the inkjet printing technology realized fluid deposition of 0.6 pl . A droplet of ∼10 μm diameter will be created if dye solution of 0.6 pl is deposited into an elastomer. Such a small droplet will exhibit WGM resonance with higher quality factor. (The quality factors of the current droplets were ∼103.) An inkjet printer can also produce a droplet array, which will exhibit attractive optical functions due to interaction between the droplets.
A microdroplet of dye solution was formed in an elastomer by using a needle-dipping method. When the droplet was excited by pulsed green laser, WGM emission was observed in the 570–610 nm wavelength range. As the elastomer or the droplet deformed, the resonance peaks shifted toward longer wavelengths. This droplet-elastomer compound will be useful to fabricate a tunable microlaser, since it enhances stability and handling-capability of droplets without spoiling its deformability or resonance tunability.
References and links
1. C. G. B. Garret, W. Kaiser, and W. L. Bond, “Stimulated emission into optical whispering modes of spheres,” Phys. Rev. 124, 1807–1809 (1961). [CrossRef]
2. R. E. Benner, P. W. Barber, J. F. Owen, and R. K. Chang, “Observation of structure resonance in the fluorescence spectra from microspheres,” Phys. Rev. Lett. 44, 475–478 (1980). [CrossRef]
3. H.-B. Lin, J. D. Eversole, and A. J. Campillo, “Spectral properties of lasing microdroplets,“ J. Opt. Soc. Am. B 9, 43–50 (1992). [CrossRef]
4. K. Sasaki, H. Fujiwara, and H. Masuhara, “Photon tunneling from an optically manipulated microsphere to a surface by lasing spectral analysis,” Appl. Phys. Lett. 70, 2647–2649 (1997). [CrossRef]
5. 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]
7. A. Kiraz, S. Ç. Yavuz, Y. Karadağ, A. Kurt, A. Sennaroglu, and H. Çankaya, “Large spectral tuning of liquid microdroplets standing on a superhydrophobic surface using optical scattering force,” Appl. Phys. Lett. 91, 231102–1–3 (2007). [CrossRef]
8. H. Finkelmann, S. T. Kim, A. Muñoz, P. Palffy-Muhoray, and B. Taheri, “Tunable mirrorless lasing in cholesteric liquid crystalline elastomers,” Adv. Mater. 13, 1069–1072 (2001). [CrossRef]
9. M. R. Weinberger, G. Langer, A. Pogantsch, A. Haase, E. Zojer, and W. Kern, “Continuously color-tunable rubber laser,” Adv. Mater. 16, 130–133 (2004). [CrossRef]
12. I. Teraoka and S. Arnold, “Enhancing the sensitivity of a whispering-gallery mode microsphere sensor by a high-refractive-index surface layer,” J. Opt. Soc. Am. B 23, 1434–1441 (2006). [CrossRef]
13. Sonoplot, Inc., “GIX Microplotter,” http://www.sonoplot.com/products.html.