Multilayer nano-particle image velocimetry

Nano-particle image velocimetry (nPIV), based on evanescent-wave illumination of fluorescent colloidal tracers, measures the two velocity components parallel to the wall averaged over the first few hundred nanometers next to the wall. The intensity of the evanescent wave decays exponentially with z, or the distance normal to the wall. Illuminated tracers closer to the wall therefore have images that are brighter than those farther from the wall. This nonuniform illumination presents the possibility to extend the technique to "multilayer nPIV," where the two velocity components parallel to the wall can be estimated at different z-locations within the illuminated region. In this paper, the variation of tracer image intensity with distance from the wall was predicted using diffraction optics-based approaches. The predictions, which were validated by calibration experiments, show that particle image intensity decays exponentially with distance normal to the wall. The feasibility of multilayer nPIV was evaluated using artificial images of plane Couette flow that incorporate evanescent-wave illumination, hindered Brownian diffusion and image noise. Each image was divided into three sub-images based on tracer image intensity, and standard techniques were then used to extract temporally and spatially averaged velocities at three different z-locations. In these simulations, velocity data were obtained within 80 nm of the wall, a threefold improvement over previous measurements. The results demonstrate that multilayer nPIV is feasible if appropriate classification techniques are developed and used to separate tracer images into different layers.