A light-driven micromanipulation system with real-time user-feedback control is used to simultaneously trap colloidal suspensions enabling a unique interactive sorting capability and arbitrary patterning of microscopic particles. The technique is based on a straightforward phase-to-intensity conversion generating multiple beam patterns for manipulation of particles in the observation plane of a microscope. Encoding of phase patterns in a spatial light modulator, which is directly controlled by a computer, allows for dynamic reconfiguration of the trapping patterns, where independent control of the position, size, shape and intensity of each beam is possible. Efficient sorting of microsphere mixtures of distinct sizes and colors using multiple optical traps is demonstrated.
© 2002 Optical Society of America
User-interactive observation and light-driven manipulation of multiple particles in a colloidal suspension is a powerful tool that can be applied in a number of applications within life and physical sciences. Light-driven manipulation works on the transfer of momentum of focused laser light to facilitate optical trapping and non-invasive control of a microscopic particle [1, 2]. Hence, inducing multiple foci or trapping beams along the observation plane allows for multiple particles to be trapped, observed and manipulated in parallel. The collective utility of dynamic multiple beams can be an indispensable tool for construction and powering of micromachines , microfluidic gears, pumps and valves [4, 5]; guided assembly of colloidal crystals [6, 7]; and formation of polymeric photonic wires . Interactive optical micromanipulation can also be integrated in miniaturized systems utilizing microfluidics and ‘lab-on-a-chip’ technology [9 – 12] enabling precise sorting and analysis of specific particles or biological cells in a self-contained sample volume.
In microfluidic applications, optical trapping has assisted in the fabrication of linear colloidal structures by laser-induced polymerization . Alternatively, this procedure can be flexibly implemented by directly maneuvering multiple trapping beams to guide an assortment of particles forming various patterns along the trapping plane. The particles can be controlled arbitrarily prior to the polymerization process resulting in the assembly of a variety of colloidal structures, which are not limited to linear configurations. Moreover, as a tool for biologists, a multiple-beam trapping system can be used for supervised or unsupervised cell sorting mechanisms while the cells are observed under the microscope. Sorting of cells that naturally exist in different sizes, shapes and other properties can be easily managed if each trapping beam has the capability to mimic the profile of particular cells.
Several optical trapping methods have been demonstrated to suit different applications, for instance, the use of two counter-propagating laser sources for single-cell manipulation and sorting . Multiple optical traps using scanning mirrors  and intensity patterns generated by interferometry  or by holographic reconstruction  have also been implemented for trapping a plurality of particles. Recently, we have demonstrated a straightforward phase-to-intensity conversion applying the Generalized Phase Contrast (GPC) method  for generating multiple beam patterns [18, 19] for simultaneous trapping and dynamic rotation of particles in the observation plane of a microscope. In this technique, the intensity distribution corresponding to the trapping beams is a direct map of an input phase pattern encoded on a programmable spatial light modulator (SLM). Since the mapping is performed by a direct imaging operation, input phase patterns can be arbitrarily reconfigured using a computer graphical user interface without the need for advanced computational algorithms. Thus, the system provides a direct-write method for producing steerable trapping beams at multiple transverse positions and allows for independent control over the shape, size and strength of each individual beam with minor loss of power.
In our previous work, we demonstrated a dynamic multiple trapping beam system by manipulating a collection of homogeneous microspheres resulting in smooth, non-discrete and pre-determined motions of the particles . In this work we show real-time, fully user-defined, arbitrary movements of optically trapped colloidal particles of different properties. User-defined multiple traps is a requirement for achieving intelligent feedback control radically advancing the concept of optical micromanipulation. Our experimental results point to practical utilization of our system in user-interactive sorting of inhomogeneous particle mixtures dispersed in water solution. To our knowledge this is the first demonstration of real-time segregation of particles according to distinct physical properties using multiple trapping beams.
2. Optical setup for generating multiple traps
Figure 1(a) shows the schematic of the interactive light-driven manipulation system as a simple optical attachment to a commercially available inverted microscope. A collimated beam from a 200-mW diode laser operating at 830-nm-wavelength is incident on a reflection-type phase-only SLM (Hamamatsu Photonics) on which patterns are encoded by a computer. The reflected light transports the encoded phase patterns through a phase contrast imaging system (lenses L1 and L2 with a π-shift phase contrast filter PCF in the Fourier plane) resulting in a high efficiency phase-to-intensity conversion at the image plane (IP). Lens L3 and a ×100 oil immersion microscope objective (MO, Numerical Aperture 1.4) form a telescopic system, which scales the intensity distribution from the plane IP onto the observation plane of the microscope. The fluorescence port of the microscope is used to couple the infrared light to the back-focal plane of the MO via a dichroic mirror (DM). The trapped particles are monitored using the standard functionality of the microscope and a CCD camera. Figures 1(b – d) show the conversion of various input phase patterns into corresponding high-contrast intensity distributions taken at the image plane IP. It demonstrates the effective generation of regular and irregular geometries and arbitrary arrays of multiple trapping beams. Figure 1(c) shows an array of 10×10 traps with each beam having a dark focus . These beams are used to trap particles with refractive indices lower than that of the surrounding medium. Figure 1(d) shows an asymmetric array of 15×15 traps with some of the traps reduced in size.
3. Light-driven manipulation of microscopic particles
3.1 Controlling unwanted stacking of particles
In our optical trapping experiments, polystyrene microspheres in an aqueous solution are lifted-off the bottom surface of a glass chamber (depth ≈ 30 μm) due to radiation pressure. Since the scattering force of the beam just exceeds the weight of the particle and the drag force of the surrounding medium, the beam will levitate the particle along the optical axis. Thus, trapped particles are approximately coplanar just below the upper surface of the glass chamber. Additional particles at the bottom may also align underneath a previously trapped particle. This has been the basis of stacking particles to form linear arrays along the beam axis [15, 21, 22]. Experiments with Bessel-beams were also recently shown to trap, stack, and rotate particles simultaneously in multiple planes . However, if the intention is to sort these particles according to type, stacking of particles oriented along each beam axis is undesirable since particles in the same stack can have different physical properties.
Figure 2 illustrates how our system can efficiently deal with this problem. In our experiments, stacked polystyrene microspheres (Bangs Laboratories, Fishers, IN) are easily recognized because the corresponding image appears blurred and distinguishable from the images of singly trapped particles (Fig. 2a). By a simple “click-and-drag” procedure on the computer, the beam associated with the stacked microspheres is selected and transversely displaced to remove the excess particle (Fig. 2b, c) while keeping the other particles trapped. An additional trapping beam is then immediately introduced at the position of the excess 2-μm-diameter bead that was removed from the stack leading to a situation where each microsphere is exclusively trapped in a corresponding beam (Fig. 2d). In contrast to a collective mechanical shaking procedure, this interactive optical procedure isolates the disturbance to the trapping beam that contains excess particles, avoiding the possible ejection of other particles from their traps. This technique of eliminating stacks of particles is easily implemented in the setup regardless of the size and the new position of the excess bead. The ability to remove particle stacking does not compromise trapping performance, keeping a configuration of particles having different sizes with one trapping beam apiece (Fig. 2d). This feature, in conjunction with the dynamic and parallel motion of the beams, makes the system a viable tool for interactive particle sorting.
3.2 Interactive sorting of inhomogeneous mixtures
We carried out particle sorting experiments by interactive optical manipulation of inhomogeneous mixtures of microspheres. Figure 3 shows an image sequence of four 2-μm-diameter and three 5-μm-diameter polystyrene spheres initially trapped at different positions by corresponding beams. The computerized graphical user interface is used to directly maneuver the traps resulting into the subsequent sorting of the beads according to size. The formation of layers separating the two distinct sizes of microspheres is seen in the last frame.
In another demonstration shown in Fig. 4, 3-μm-diameter dyed polystyrene spheres (Polysciences, Inc.) were trapped by multiple beams and then sorted on the basis of color. Triplets of blue, red and yellow microspheres are sorted into three separate layers. The two sorting experiments in Fig. 3 and Fig. 4 illustrate that the trapping beams in the system have advanced particle maneuverability to simultaneously direct particles along arbitrary paths on the trapping plane, which would be difficult to achieve, if not impossible, with other techniques.
Manipulation of particles with discernible properties that can be viewed directly in the observation plane makes it straightforward to apply pattern recognition or vision tracking functionalities providing researchers a practical experimental tool across a broad range of applications. To our knowledge results in Fig. 3 and Fig. 4 are the first demonstrations of interactive sorting of microscopic particles using parallel multiple-beam manipulation. For the sorting experiments, the average speed at which we can move a microsphere with diameters considered is approximately 2 μm/s. This sorting capability can be applied to light-assisted formation of colloidal microspheres useful for microfluidic systems. However, due to limited laser power of our current setup, actual manipulation of larger arrays of particles with high-speed actuation is still not achievable. Nevertheless, this experimental demonstration shows a proof-of-principle for real-time simultaneous trapping and interactive sorting of micron-sized particles. Applying a stronger laser source will allow faster dynamics of large arrays of particles in conjunction with increased positional accuracy due to the reduced influence of Brownian motion.
The inherent adaptability of the system to create a variety of trapping beam patterns can potentially enhance, and even combine altogether, current optical micromanipulation schemes such as the rotation of irregularly shaped objects with rectangular beams , trapping of low-index particles by hollow beams , and controlled trapping of a single particle surrounded by multiple beams . Furthermore, the approach is straightforward because a desired intensity distribution can be generated using a graphical user interface that rapidly encodes the corresponding phase pattern on the SLM.
Multiple beams generated by the system can be utilized, not only for optically assisted synthesis of functional microstructures, but also for non-contact and parallel actuation of these devices. The system we describe here can provide flexible control over an assembly of micro-devices, crucial for advanced light-powered and controlled ‘lab-on-a-chip’ demonstrations in the future. Figure 5 shows a hypothetical computer-programmable ‘lab-on-a-chip’ system with microfluidic elements such as pumps and valves at the compartments and junctions of the channels. The elements are simultaneously assembled and actuated by multiple optical traps to facilitate the segregation of particles with different properties. Arbitrarily shaped beam configurations are also possible such as hollow beams for trapping of low-index particles and grid patterns for guided assembly of colloidal crystals.
We have demonstrated an interactive particle-sorting instrument for microscopic particles in a colloidal suspension. To our knowledge, this is the first demonstration of a user-controlled sorting of particle mixtures using multiple optical traps. The system is based on the generation of multiple trapping beams in the sample region where particles are manipulated. Arbitrarily shaped trapping beam configurations are obtained from an optical setup that performs efficient conversion of phase patterns into corresponding intensity distributions. The spatial distribution of trapping beams projected onto the microscope observation plane is a direct map of the phase pattern encoded on a programmable spatial light modulator. Finally, the multiple trapping beams can be used to assemble and actuate versatile microstructures for a range of microfluidic functionalities.
We thank the Danish Technical Scientific Research Council for supporting this research and T. Hara and Y. Kobayashi of Hamamatsu Photonics for useful discussions on the operation of the SLM.
References and links
1. A. Ashikin, J. M. Dziedzic, J. E. Bjorkholm, and S. Chu, “Observation of a single-beam gradient force optical trap for dielectric particles,” Opt. Lett. 11, 288 (1986). [CrossRef]
3. M. E. J. Friese, H. Rubinsztein-Dunlop, J. Gold, P. Hagberg, and D. Hanstrop, “Optically driven micromachine elements,” Appl. Phys. Lett. 78, 547 (2001). [CrossRef]
5. A. Terray, J. Oakley, and D. W. M. Marr, “Fabrication of linear colloidal structures for microfluidic applications,” Appl. Phys. Lett. 81, 1555 (2002). [CrossRef]
8. S. M. Mahurinet al., “Photonic polymers: a new class of photonic wire structure from intersecting polymer-blend microspheres,” Opt. Lett. 27, 610 (2002). [CrossRef]
11. A. Y. Fu, C. Spence, A. Scherer, F. H. Arnold, and S. R. Quake, “A microfabricated fluorescence-activated cell sorter,” Nature Biotechnol. 17, 1109 (1999). [CrossRef]
12. T. Mülleret al., “A 3D-micro electrode for handling and caging single cells and particles,” Biosensors Bioelectronics 14, 247 (1999). [CrossRef]
13. S. C. Grover, A. G. Skirtach, R. C. Gauthier, and C. P. Grover, “Automated single-cell sorting system based on optical trapping,” J. Biomedical Opt. 6, 14 (2001). [CrossRef]
14. K. Sasaki, M. Koshioka, H. Misawa, N. Kitamura, and H. Masuhara, “Pattern formation and flow control of fine particles by laser-scanning micromanipulation,” Opt. Lett. 16, 1463 (1991). [CrossRef] [PubMed]
16. M. Reicherter, T. Haist, E. U. Wagemann, and H. J. Tiziani, ”Optical particle trapping with computer-generated holograms written on a liquid-crystal display,” Opt. Lett. 24, 608 (1999). [CrossRef]
17. J. Glückstad and P. C. Mogensen, “Optimal phase contrast in common-path interferometry,” Appl. Opt. 40, 268 (2001). [CrossRef]
18. R. L. Eriksen, P. C. Mogensen, and J. Glückstad, “Multiple-beam optical tweezers generated by the generalized phase-contrast method,” Opt. Lett. 27, 267 (2002). [CrossRef]
19. R. L. Eriksen, V. R. Daria, and J. Glückstad, “Fully dynamic multiple-beam optical tweezers,” Opt. Express 10, 597 (2002), http://www.opticsexpress.org/abstract.cfm?URI=OPEX-10-14-597. [CrossRef] [PubMed]
20. J. Arlt and M. Padgett, “Generation of a beam with a dark focus surrounded by regions of higher intensity: the optical bottle beam,” Opt. Lett. 25, 191 (2000). [CrossRef]
21. P. Zemanek, A. Jonas, L. Sramek, and M. Liska, “Optical trapping of nanoparticles and microparticles by a Gaussian standing wave,” Opt. Lett. 24, 1448 (1999). [CrossRef]
22. J. Arlt, V. Garcés-Chávez, W. Sibbett, and K. Dholakia, “Optical micromanipulation using a Bessel light beam,” Opt. Commun. 197, 239 (2002). [CrossRef]
23. V. Garcés-Chávez, D. McGloin, H. Melville, W. Sibbett, and K. Dholakia, “Simultaneous micromanipulation in multiple planes using a self-reconstructing light beam,” Nature 419, 145 (2002). [CrossRef] [PubMed]
24. A. O’Neil and M. Padgett, “Rotational control within optical tweezers by use of a rotating aperture,” Opt. Lett. 27, 743 (2002). [CrossRef]
26. Y. Ogura, N. Shirai, and J. Tanida, “Optical levitation and translation of a microscopic particle by use of multiple beams generated by vertical-cavity surface-emitting laser array sources,” Appl. Opt. 41, 5645 (2002). [CrossRef] [PubMed]