An experimentally-achieved information-driven Brownian motor shows maximum power at the relaxation time

We present an experimental realization of an information-driven Brownian motor by periodically cooling a Brownian particle trapped in a harmonic potential connected to a single heat bath, where cooling is carried out by the information process consisting of measurement and feedback control. We show that the random motion of the particle is rectified by symmetry-broken feedback cooling where the particle is cooled only when it resides on the specific side of the potential center at the instant of measurement. Studying how the motor thermodynamics depends on cycle period τ relative to the relaxation time τ_B of the Brownian particle, we find that the ratcheting of thermal noise produces the maximum work extraction when τ ≥ 5τ_B, while the extracted power is maximum near τ = τ_B, implying the optimal operating time for the ratcheting process. In addition, we find that the average transport velocity is monotonically decreased as τ increases and present the upper bound for the velocity.