Figure 1

(Color) Magnetoresistivity traces of the antidot lattice (blue) and intact 2DEG part (red). An atomic force microscope image of the lattice including the etched profile (with sketch of the commensurability orbits 3 and 4) as well as the electron system parameters are shown in the two insets.Reuse & Permissions

Figure 2

MW irradiation setup. The direction of the MW electric field in the waveguide is indicated. Dashed arrows show the propagation direction of MWs. The sample can be turned by $90°$ in the plane.Reuse & Permissions

Figure 3

(Color) Magnetic-field dependence of the dc voltage $\left(U_A-U_B\right)$ measured in “ratchet” antidot lattice under MW irradiation having the vector of linear polarization parallel to the $x$ axis (red) and perpendicular to it (blue). Importantly, the sign of the signal is opposite for the two orientations of polarization at zero magnetic field. At higher magnetic fields $\left(B>0.2\text{T}\right)$, the effect vanishes for any direction of MW polarization. The absence of the “ratchet” signal in a lattice of symmetrical circular scatterers (orange and violet) proves that broken symmetry is a necessary ingredient to the “ratchet” effect. The insets show the power dependence of the polarization-dependent signal (to the right) and the measurement configuration with a schematic representation of the scattering process (on one antidot) for both directions of polarization (to the left).Reuse & Permissions

Figure 4

Magnetic-field dependence of the photovoltage measured in another sample for both directions of polarization at 44.35 GHz.Reuse & Permissions

Figure 5

(Color) Temperature dependence of the “ratchet” signal (at zero magnetic field) plotted together with temperature dependencies of the mobility in the sample for “ratchet” and unpatterned 2DEG (temperature and mobility axes are presented in logarithmical scales).Reuse & Permissions