Figure 2: Chirality effects in a vortex flow.
(a) The mean orbit radii 
observed in experiments depends on the chirality of the particles and on the direction of flow (solid markers=CW cell rotation; open markers=CCW rotation). For all cell velocities ωcell, 
and 
. Upon ‘flow reversal’ (without any manipulation of particles), 
and 
. Error bars correspond to standard deviations in 
, which are based on at least 10 independent experiments (12,500 data points per run recorded) for each value of ωcell (note: four particles of each type were tested). (b) Scheme illustrating why the behaviour of a chiral particle in a CW flow is expected to be equivalent to its ‘mirror image’ in a CCW flow. Rotation of the entire system by 180° about a horizontal axis (denoted by grey arrows) changes the particle chirality and the flow direction (with respect to an observer looking down the vertical axis as denoted by the eye cartoon) while the orbit radius remains unchanged. These symmetry arguments are in agreement with the experimental observations in a. (c) When the direction of flow reverses from CW to CCW, the linear and angular velocities, ω and vθ, also change as illustrated by the bar graph showing the ratios of the particle velocities for CW and CCW flows averaged over all experiments. Error bars represent s.d. based on at least 200 experiments for each bar. (d) The average angular velocity of the particles 
is consistently smaller than that of the Couette cell ωcell. Error bars represent s.d. based on at least 200 experiments for each bar.
