Open Access
Add to BibSonomyAbstract
We exploit a fiber puller to transform a telecom single-mode optical fiber with a 125 µm diameter into a symmetric and unbroken slightly tapered optical fiber with a 50 µm diameter at the minimum waist. When the laser light is launched into the optical fiber, we can observe that, due to the evanescent wave of the slightly tapered fiber, the nearby polystyrene microparticles with 10 µm diameters will be attracted onto the fiber surface and roll separately in the direction of light propagation. We have also simulated and compared the optical propulsion effects on the microparticles when the laser light is launched into a slightly tapered fiber and a heavily tapered (subwavelength) fiber, respectively.
©2010 Optical Society of America
1. Introduction
Compared with the traditional optical tweezer [1], the technique of using the evanescent wave out of the optical waveguide to attract and transport the microparticles has a longer controllable delivery range and a larger structural variety in the manipulation process. The optical manipulation of microspheres with evanescent fields was first reported in 1992 by using a high refractive index prism [2]. Subsequently, various types of optical waveguides were investigated for the application in microparticle optical propulsion [3,4]. Although a submicrometer high refractive index waveguide can generate high intensity evanescent fields and enhance the optical propelling of microparticles with higher propulsion velocities [5], the fabrication process of a submicrometer strip waveguide is very sophisticated. The optical propulsions of microspheres along subwavelength optical wires that were manufactured from single-mode optical fibers by the flame-brushing technique have also been demonstrated recently [6,7]. A short section of the optical fiber was tapered to submicrometer dimensions such that a considerable fraction of the propagating mode can lie outside the boundary of the optical fiber and thus can be exploited for the optical trapping and propulsion of microparticles. It has the benefit of flexibility in a 3D geometry and a very low optical insertion loss compared with the planar waveguide. Nevertheless, the preparation of the optical fiber nanowires or subwavelength (heavily tapered) optical wires needs very careful handling, bringing about many difficulties during the manipulation. Here, we report the optical propulsion of microparticles using the evanescent wave of a slightly tapered optical fiber, which is easily manufactured. The optical propulsion effects on the microparticles of a slightly tapered fiber and a heavily tapered fiber are also simulated and compared.
2. Experimental system and methods
We first exploit a fiber puller (P-2000, SUTTER INSTRUMENT CO., heating source: CO2 Laser) to transform a telecom single-mode optical fiber (POFC SMF 130V, Prime Optical Fiber Corp., cut-off wavelength at 1150~1330 nm) [Fig. 1(a) ] into a symmetric and unbroken slightly tapered optical fiber [Fig. 1(b)]. The diameter of the original telecom single-mode optical fiber is ~125 µm and the minimum waist diameter of the slightly tapered optical fiber is measured to be ~50 µm by using an extra optical microscope.
Fig. 1 The observed optical microscopic images of (a) the original telecom single-mode optical fiber (diameter ~125 µm) and (b) the slightly tapered optical fiber (minimum waist diameter ~50 µm). The observed optical patterns of the scattered light around the slightly tapered optical fiber when the fiber output power is (c) 10 mW and (d) 30 mW, respectively.
The experimental setup for trapping and propelling microparticles by a slightly tapered optical fiber is shown in Fig. 2 . The front end of the optical fiber (untapered region) is fixed by a fiber holder and the rear end is supported by an adjustable diaphragm. The fiber taper region is positioned in a microfluidic channel on a microscope slide which is formed by two adhesive tapes. A 980 nm diode laser is coupled into the slightly tapered optical fiber through a 20X microscope objective lens. Then we use a 50X long-working-distance microscope objective lens and a CCD camera to observe the optical microscopic image of the slightly tapered fiber from the top. The major illumination is from the white light LED at below. When an optical filter is not placed in front of the CCD camera, we can observe the optical patterns of the scattered light emanating from the slightly tapered fiber due to the evanescent wave around the fiber surface when the fiber output power is 10 mW [Fig. 1(c)] and when the fiber output power is 30 mW [Fig. 1(d)]. The scattered light intensity becomes stronger as the laser power launched into the tapered fiber increases. In the experiments, when the drive current of the 980 nm laser diode is set at 205 mA, the delivered optical power of the laser diode is ~100 mW. Under this condition, the optical power of the laser light coupled into the input end of the slightly tapered fiber is estimated to be ~30 mW in a conservative level by our experimental experience, and the optical power from the fiber output end is measured to be 10 mW, which denotes an optical loss of the fiber tapers at ~4.8 dB.
Fig. 2 The (a) configuration and (b) photograph of the experimental setup for trapping and propelling microparticles by a slightly tapered optical fiber. MO, microscope objective lens; PC, personal computer; CCD, charge coupled device; LD, laser diode; LED, light emitting diode.
Then we perform the experiments on the microparticle manipulation. The polystyrene microparticles are diluted in the de-ionized water, and several drops of the suspension are laid on the fiber taper section which is positioned in a microfluidic channel on a microscope slide with a pipette. Through the microscope imaging system which takes an advantage of an optical filter to block off most of the laser light, we can observed that, due to the evanescent wave around the slightly tapered fiber, the microparticles will be attracted and trapped on the fiber surface when the microparticles flow near the slightly tapered fiber, and then the microparticles will be propelled by the radiation pressure and roll along the slightly tapered fiber in the direction of light propagation [6,7]. Because the viscous drag force of the surrounding water suspension will counteract with the optical propulsion force, the microparticle can be only accelerated up to a constant terminal speed at balance.
3. Experimental results
We first observe the phenomena of the optical attraction and transportation of a single polystyrene microparticle with a 10 µm diameter on the surface of the slightly tapered optical fiber in the direction of light propagation. The 980 nm laser light is launched into the input end of the slightly tapered optical fiber, and the fiber output power is measured to be 10 mW. The microscopic images in Fig. 3 show that, within an interval of 4 s, the single microparticle B has moved a distance of ~25 µm. The average propulsion velocity of the microparticle B is thus calculated to be 6.25 µm/s. The microparticle A is stuck to the fiber surface by a certain cause, such as the resultant strong viscosity of a deteriorated polystyrene microparticle. Hence, the microparticle A cannot move and has a zero propulsion velocity.
Fig. 3 The observed microscopic images of the transportation of a single microparticle on the surface of the slightly tapered optical fiber when the fiber output power is 10 mW.
We have also observed the optical attraction and transportation of two separate polystyrene microparticles. When the fiber output power is 10 mW, Figs. 4(a) and 4(b) show that, within an interval of 4 s, the microparticle C has moved a distance of ~13.5 µm and its propulsion velocity is 3.38 µm/s; the microparticle D has moved a distance of ~23.1 µm and its propulsion velocity is 5.78 µm/s. When the fiber output power is increased to 20 mW, Figs. 4(c) and 4(d) show that, within an interval of 6.8 s, the microparticle E has moved a distance of ~25.1 µm and its propulsion velocity is 3.69 µm/s; the microparticle F has moved a distance of ~40.4 µm and its propulsion velocity is 5.94 µm/s. The propulsion velocity of the microparticle is found to be only slightly enhanced by increasing the launched laser power twice. The marginal increase in the propulsion velocity might be due to the nonlinear dependence of the viscous force of the water suspension and the friction force of the stripped fiber surface on the microparticle velocity in our system. As to the difference between the velocities of the microparticles on the two sides of the tapered fiber, we attribute it to the inhomogeneous smoothness on the surface of the stripped fiber, which is manifested implicitly by the scattered optical patterns shown in the Figs. 1(c) and 1(d). Besides, the bright interference fringes on the trapped microparticles in Figs. 4(c) and 4(d) are attributed to the whispering gallery modes excited in the microparticles by the laser light [6,7].
Fig. 4 The observed microscopic images of the transportation of two separate microparticles on the surface of the slightly tapered optical fiber when the fiber output power is (a)-(b) 10 mW (Media 1), and (c)-(d) 20 mW (Media 2), respectively.
4. Simulation results and discussion
Moreover, we use a software (COMSOL Multiphysics) to simulate the phenomena observed in the experiment by the finite element method. We first simulate the propagation and variation of the laser light that is launched into a slightly tapered fiber and observe the optical influences of the evanescent wave on the dielectric microparticle at the nearby surface of the slightly tapered fiber. In order to explore the overall situations in various sections of the slightly tapered fiber and overcome the limitations of the data storage and processing capability of the personal computer, we reduce the size of the slightly tapered fiber in the simulation to be fifth of that in the experiment. We assume that the original optical fiber has a core diameter of 1.65 µm and a cladding diameter of 25 µm. The slightly tapered optical fiber has a core diameter of 0.66 µm and a cladding diameter of 10 µm in the minimum waist section. The step-index optical fiber used in the simulation has a core index n1 = 1.4457 and a cladding index n2 = 1.4378. The dielectric microparticle has a diameter of 10 µm and a refractive index n3 = 1.574. The surrounding medium (water) has a refractive index n4 = 1.33. The optical wave mode is considered as a plane TE mode and the optical wavelength is assumed to be 980 nm.
The simulation results are shown in Fig. 5 and reveal that the incident optical wave (from the left side) in the fiber core will be massively coupled into the fiber cladding at the taper transition section, therefore the evanescent wave around the surface of the slightly tapered fiber will afford to attract the microparticle via the gradient force at various sections of the slightly tapered fiber. The coupled laser light within the microparticle will hit and propel the microparticle at the right-down side boundary [Figs. 5(a)–5(c)] via the radiation pressure, and thus can transport the microparticle to roll on the surface of the slightly tapered fiber in the direction of light propagation [6,7]. The optical power coupled into the microparticle will decay gradually when the microparticle is situated toward the end section of the slightly tapered fiber [Fig. 5(d)] where the evanescent wave becomes somewhat weaker. In the experiment, we have also ever observed the transportation of microparticles on the surface of the slightly tapered fiber at the taper transition sections, where the evanescent wave is still capable of trapping and propelling the microparticles. This small difference between the experiments and the simulation results might be caused by the scale reducing of the simulation parameters.
Fig. 5 Calculated light wave distributions of a slightly tapered optical fiber and a dielectric microparticle as the longitudinal relative position of the microparticle is set at (a) −15 µm, (b) −7 µm, (c) 0 µm, (d) 7 µm, respectively.
We next shrink the cladding diameter in the minimum waist of the tapered fiber to be 1 µm to strengthen the degree of fiber finesse, and change the optical wavelength of the input laser light to be 1.5 µm to simulate the optical distributions at the interior and the surrounding of a subwavelength heavily tapered fiber. A dielectric microparticle with a 5 µm diameter is placed nearby the surface of the hyperfine tapered fiber at various longitudinal positions. The optical influences of the evanescent wave on the microparticle are simulated and shown in Fig. 6 . The light wave coupled into the microparticle through the evanescent wave of a heavily tapered fiber (Fig. 6) looks much brighter than that in the case of a slightly tapered fiber (Fig. 5) because the evanescent wave around a subwavelength tapered fiber is much stronger [6,7]. On the other hand, the coupled light wave within the microparticle will also hit and propel the microparticle at the right-down side boundary [Figs. 6(a) and 6(b)] via the radiation pressure, and thus can also transport the microparticle to roll on the surface of the heavily tapered fiber in the direction of light propagation. Nevertheless, the power fraction of the light wave coupled into the microparticle is so high such that the remaining light in the subwavelength heavily tapered fiber behind the microparticle is quite weak and cannot support the attraction and transportation of another microparticle separately. As a result, the subwavelength heavily tapered fiber (optical wire) can only trap and propel a single microparticle or a cluster of several microparticles, as demonstrated in Ref [7]. This could be a drawback of the subwavelength heavily tapered fiber in the application of many-particle separate manipulation. Besides, the light power coupled into the microparticle will decay massively when the microparticle is situated toward the end section of the subwavelength heavily tapered fiber [Fig. 6(c)] where the evanescent wave becomes also much weaker.
Fig. 6 Calculated light wave distributions of a subwavelength heavily tapered optical fiber and a dielectric microparticle as the longitudinal relative position of the microparticle is set at (a) −7 µm, (b) 0 µm, (c) 7 µm, respectively.
5. Conclusion
We have successfully demonstrated the optical attraction and transportation of one or two dielectric microparticles separately on the specific fiber surface through the evanescent wave of a slightly tapered optical fiber, which is easily manufactured. The optical fiber sensor based on a slightly tapered optical fiber and micro-fluidics will find great usage in the application of microparticle or biological cell manipulation. The simulation results can elucidate the optical phenomena observed in the experiments. The case of a subwavelength heavily tapered optical fiber is also simulated, and we can observe a much brighter evanescent wave and a much stronger optical coupling or propelling effect. Yet, the simulation also reveals the disadvantage of a hyperfine tapered fiber in the application of many-particle separate transportation. A slightly tapered optical fiber is shown to be a more excellent and convenient tool for optical propulsion of microparticles by the evanescent wave in some aspects, unless we would like to achieve the optical manipulation of nanoparticles.
Acknowledgments
We acknowledge the financial support from the National Science Council, Taiwan, through Project NSC 97-2112-M-415-002-MY3.
References and links
1. A. Ashkin, “Acceleration and Trapping of Particles by Radiation Pressure,” Phys. Rev. Lett. 24(4), 156–159 (1970). [CrossRef]
2. S. Kawata and T. Sugiura, “Movement of micrometer-sized particles in the evanescent field of a laser beam,” Opt. Lett. 17(11), 772–774 (1992). [CrossRef] [PubMed]
3. B. S. Schmidt, A. H. Yang, D. Erickson, and M. Lipson, “Optofluidic trapping and transport on solid core waveguides within a microfluidic device,” Opt. Express 15(22), 14322–14334 (2007). [CrossRef] [PubMed]
4. A. H. J. Yang, S. D. Moore, B. S. Schmidt, M. Klug, M. Lipson, and D. Erickson, “Optical manipulation of nanoparticles and biomolecules in sub-wavelength slot waveguides,” Nature 457(7225), 71–75 (2009). [CrossRef] [PubMed]
5. B. S. Ahluwalia, A. Z. Subramanian, O. G. Hellso, N. M. B. Perney, N. P. Sessions, and J. S. Wilkinson, “Fabrication of Submicrometer High Refractive Index Tantalum Pentoxide Waveguides for Optical Propulsion of Microparticles,” IEEE Photon. Technol. Lett. 21(19), 1408–1410 (2009). [CrossRef]
6. G. Brambilla, G. S. Murugan, J. S. Wilkinson, and D. J. Richardson, “Optical manipulation of microspheres along a subwavelength optical wire,” Opt. Lett. 32(20), 3041–3043 (2007). [CrossRef] [PubMed]
7. G. S. Murugan, G. Brambilla, J. S. Wilkinson, and D. J. Richardson, “Optical Propulsion of Individual and Clustered Microspheres along Sub-Micron Optical Wires,” Jpn. J. Appl. Phys. 47(8), 6716–6718 (2008). [CrossRef]
