Fig. 1. The experimental optofluidic geometry, comprising a pair of waveguides separated by a microfluidic channel. A probe beam propagates between the waveguides, being influenced by the contents of the channel. In this paper we utilize standard single mode optical fiber for the waveguides.

Fig. 2. Schematic process diagram for the fabrication of the semi-planar buried fiber device. a) The SMF is suspended taut 300 μm above a glass substrate. b) A layer of photo-polymer (NOA-16) is poured over the substrate (and SMF), allowed to settle under gravity then cured. As a result, the SMF is buried some 300 μm below the surface of the polymer. c) A channel is cut in the hard photo-polymer with a dicing saw, cleaving the fiber and giving the optofluidic geometry shown in Fig. 1.

Fig. 3. A schematic of the bulk optics used to deliver the trapping beam to the buried fiber device. The beam is generated by the Q-switched vanadate laser at right. The beam is brought to the sample using an inverted microscope setup and is focused using a 40X, 0.65 NA microscope objective. The sample is illuminated from above and examined though the inverted microscope using a CCD camera. Spatial control of the sample is delivered by a piezo actuator in the plane and a stepper motor out of the plane of the sample. The beam path is shown in yellow.

Fig. 4. A photograph and cross sectional schematics of the experimental geometry. In the photograph a micro-sphere is trapped inside the probe beam between the two buried SMFs. The horizontal red striations are from the signal beam being scattered out-of-plane by the sphere.

Fig. 5. A representative output spectrum of the device with a micro-sphere displacement of 5.8 μm. The wavelengths of the spectrum are divided into two regions: above and below the single mode cutoff of the SMF. In the multimode propagation region, the field output is highly dependent on wavelength, resulting in sharp spectral features in the output spectrum. In the single mode region, the field output is always shaped as the fundamental mode of the fiber, resulting in slowly varying spectral features.

Fig. 6. The insertion loss of the buried fiber device for various positions of the trapped micro-sphere, probed at a wavelength of 1.5 μm. Experimental data is shown with the solid line; the results of the numerical simulation are shown with the dashed line. Good agreement is seen between the experimental and numerical curves. At top the three transmission regimes are identified.

Fig. 7. FDTD field outputs for the various transmission regimes. The rectangular outline in the top center of the right hand portion of the figure shows the position and transverse extent of the output fiber. Note the different scales on each axis, resulting in the micro-sphere looking ovoid in these pictures. On the left is insertion loss curve from Fig. 7 and an arrow indicating the point on that curve the FDTD field output (at right). a) On axis transmission: the micro sphere is centered in the probe beam and is acting as a spherical lens, providing enhanced coupling. b) Off axis: the micro-sphere is steering the beam away from the core of the output SMF. c) Leaving beam: the micro-sphere is only slightly perturbing the probe beam, allowing most of the light to be collected.

Fig. 8. An animation showing the FDTD field output changing as the micro-sphere is moved across the probe beam. Notice how, as the sphere moves from the center of the beam, the light is steered away from the core of the output SMF. As the sphere is moved further out of the beam, the light returns to the core of the SMF. [Media 1]