Optical trapping, which allows trapping and manipulation of microscopic particles using light alone (“optical tweezing”), has been a boon to biologists, who can trap and sort cells, and materials scientists, who can, in just one example, move microspherical lenses for direct-write nanopatterning. But the required optical gradient forces typically are of the right magnitude only when the refractive indices of the particles to be manipulated are only modestly higher than the refractive index of the surrounding medium.
A team of scientists from the University of Colorado, the National Renewable Energy Laboratory, and the National Institute of Standards and Technology (all in Boulder, CO) and the University of Gothenburg (Göteburg, Sweden) is pursuing an approach that allows optical manipulation of microparticles with a much higher refractive index, such as gallium nitride (GaN) nanowires, which have a very high refractive index of about 2.4 for light at 587 nm.1 To achieve this, the surrounding medium is liquid crystalline (LC); the interaction of laser light with the liquid crystal (which has a relatively low index of about 1.5) leads to a number of effects that can be exploited for manipulation.
The experimental setup includes holographic optical tweezers and a fluorescence confocal polarizing microscope (FCPM). To form the optical tweezers, a 512 × 512 spatial light modulator (SLM) is imaged onto the back-aperture of the microscope objective. Two objectives were used, 60X and 100X, both with a numerical aperture of 1.42.
The GaN nanowires were fabricated using molecular-beam epitaxy, which creates wires with a typical length of about 10 μm and a hexagonal cross-section with a width of about 300 nm. Experimental LC hosts included nematic and cholesteric LCs doped with a very small amount of fluorescent dye to allow FCPM imaging without altering the properties of the LC. Optical manipulation was done at a 1064 nm wavelength, while imaging relied on excitation at 488 nm and detection in the 505–525 nm wavelength range.
Nematic LC host
Left by themselves, nanowires in a nematic LC host tend to align themselves along the LC direction (which minimizes elastic energy), with a small tilt due to influence from the surface-rubbed LC alignment layer. Focusing a low-power (less than 25 mW) laser beam near a nanowire attracts the wire to the beam; then, once the wire is trapped, it pushes the wire along the beam.
Placing the optical trap at one end of a nanowire places a torque on the wire due to scattering, which rotates the wire out of plane until the torque is balanced by the LC’s elastic torque; removing the torque causes the wire to rotate back to its initial position.
Higher optical powers (around 35 mW or more) cause the local direction of the LC itself to change, which changes the elastic forces on the nanowire. The result is a repelling force as the wire moves to minimize elastic free energy. As a result, a nanowire that stays parallel to the LC direction can be translated in a perpendicular direction using a single trap; with two traps, one stationary and one moving, the wire can be rotated (at least until the added elastic energy becomes large enough to overcome the torque).
Cholesteric LC host
In a cell with two surfaces having parallel-rubbed alignment, a cholesteric LC falls into a helical form that produces a coupling between a nanowire’s position along the z-axis and its in-plane rotation angle. As a result, as an optical trap at the center of a nanowire pushes it, the wire also rotates as it passes along the helical structure (see figure).
Conversely, two traps positioned at the nanowire’s ends that cause the wire to rotate also force the wire to move, screwlike, along the z-axis. The nanowire remains parallel in the x-y plane as it rotates, even if a trap is positioned at the end of the wire; this is due to elastic forces preventing the wire from tipping toward the z-axis. Such manipulations actually allow the researchers to determine the local “pitch,” or amount of helicity, in the cholesteric LC.
These types of nanowire manipulations also enable probing discontinuous cholesteric LC samples, with wire motions helping to locate discontinuities and characterize the twists and turns within complex defects.
REFERENCE
1. D. Engström et al., Opt. Exp., 20, 7, 7741 (2012).
