We show that all questions raised in the Comment can be easily answered within the theory formulated in the commented papers. It is also shown that the bulk theory promoted by the commenter fails to explain the most important fundamental features of the discussed phenomenon.
In the Comment1 Zudov claims that our theory 2,3does not explain the microwave induced resistance oscillations/zero resistance states (MIRO/ZRS) and that MIRO/ZRS is a bulk phenomenon. In this Reply we show that, focusing his attention on only a few very specific aspects of the effect, Zudov does not notice a number of global, fundamental problems of MIRO/ZRS, which could be understood only after assuming that this phenomenon has a near-contact origin 2,3 . The MIZRS phenomenon 4,5 has demonstrated at least five mysterious features which seemed to contradict both the previously known physics and common sense:1. The MIZRS effect is huge. The photoresistance maxima were found to be 7 − 10 times larger than the dark resistance values. The photoresistance is a nonlinear phenomenon: the dc current is proportional to the dc electric field E 0 and the squared microwave field E ω , j ∝ E dc E 2 ω . As known, substantial nonlinear effects can be observed only if the corresponding electric field parameter exceeds unity:Physically F is the additional momentum ∼ eE ω /ω (energy ∼ eE ω v F /ω), acquired by an electron during one oscillation period, normalized to its average momentum p F (energy E F ); here ω is the microwave frequency, p F , v F , E F are the Fermi momentum, velocity and energy, and n s is the density of two-dimensional (2D) electrons. The condition (1) can be rewritten in terms of the microwave power density P required for the observation of such a huge MIZRS effect:For typical MIZRS parameters (f ≃ 100 GHz, n s ≃ 3 × 10 11 cm −2 ) the required nonlinear power density (2) isIn real MIZRS experiments the microwave power density was about six orders of magnitude smaller ( 1 mW/cm 2 , Refs. 4,5 ). The first MIZRS puzzle was thus: How such a huge effect can be observed at so low microwave powers?2. The MIRO/ZRS effect demonstrates strong oscillations not only around the fundamental cyclotron frequency ω = ω c but also around harmonics ω = nω c , n = 2, 3, 4, 5, . . .. It is well known that in the uniform external ac electric field the transitions between Landau levels E N = ω c (N + 1/2) are forbidden if ∆N = ±1. In order to violate this selection rule the microwave field should be strongly inhomogeneous on the cyclotron radius (r c ) scale: the nonlocal bulk conductivity of the 2D electron gas in the magnetic field B has the formwhere X = qv F /ω c = qr c is the non-locality parameter, q is the wave-vector of the external wave and J n are Bessel functions. In the MIRO/ZRS experiments the radiation wavelength (≃ 3 mm) and the sample dimensions (≃ 0.2 − 1 mm) were much larger than the cyclotron radius (r c ≃ 1 − 5 µm). As seen from (4), the amplitude of the n-th resonance (at ω = nω c ) scales at qr c ≪ 1 as A n ∝ [n(qr c /2) n−1 ] 2