Coherent branched flow in a two-dimensional electron gas

Abstract

Images of electron flow from the quantum point contact (QPC) are obtained by raster scanning a negatively charged SPM tip above the surface of the device and simultaneously measuring the position-dependent conductance of the device. The negatively charged tip capacitively couples to the 2DEG, creating a depletion region that backscatters electron waves. When the tip is positioned over areas with high electron flow from the QPC the conductance is decreased, whereas when the tip is over areas of relatively low electron flow the conductance is unmodified. By raster scanning the tip over the sample and simultaneously recording the effect the tip has on device conductance, a two dimensional image of electron flow can be obtained.The quantum point contact sample is mounted in an atomic force microscope and cooled to liquid He temperatures. The QPC is formed in the 2DEG inside a GaAs/AlGaAs heterostructure by negatively biasing two gates on the surface -a negative potential on these gates creates two depletion regions that define a variable width channel between them as shown in Fig. 1a. The conductance of the QPC is measured using an ac lock-in amplifier at 11kHz. The heterostructure for the devices used in this experiment was grown by molecular beam epitaxy on an n-type GaAs substrate.The 2DEG resides 57 nm below the surface with mobility µ = 1.0x10 6 cm 2 /Vs and density n = 4.5x10 11 /cm 2 . These values of mobility and density correspond to a mean free path l = 11 µm, Fermi wavelength λ F = 37 nm, and Fermi energy E F = 16 meV. The root mean square voltage across the QPC was chosen so as to not heat electrons -0.2 mV for 1.7K scans. The conductance of the quantum point contact, shown in Fig. 1b, increases as the width of the channel is increased (by changing the gate voltage V g ) and shows well defined conductance plateaus at integer multiples of the conductance quantum 2e 2 /h 1,2 . When probing the electron flow, the SPM tip was held at a negative potential relative to the 2DEG and was scanned at 10nm above the surface of the heterostructure. Figures 2a and 2b show images of electron flow from two different quantum point contacts at the temperature 1.7K; both QPCs are biased on the G = 2e 2 /h conductance plateau. Figure 2b shows the flow patterns on each side of a quantum point contact (the gated region at the center was not scanned), and Figure 2a shows a higher-resolution image of flow from one side of a different QPC.In both these images, the current exits the point contact in a central lobe, as would be expected from an exact quantum-mechanical calculation of electron flow through an ideal QPC without impurities or non-uniform distributions of dopant atoms. Rather than continuing out as a smoothly widening fan, it quickly forks into several different paths and continues to branch off into ever smaller rivulets for the full width of the scan. This branching behavior was observed in all of the 13 QPC exit patterns observed so far. Previously, there have been suggestions of an unexpected narrowness in observe...

Chaos beyond linearized stability analysis: Folding of the phase space and distribution of Lyapunov exponents

“…However for Hamiltonian systems only the Lyapunov spectrum is meaningful. Its behavior at small λ, which is important for various practical applications [5,6,7,8], remains not covered in the existing literature.…”

“…Combining together eqs. (6)(7)(8) we express new coordinates u and v through initial ones u 0 and v 0…”

“…(Estimates, analogous to that of Eqs. (6)(7)(8)(9)(10)(11)(12) below, were used in Refs. [9,10] in order to describe the fractal dimension spectrum f (α) of chaotic attractor.…”

Ballistic Transport Phenomena in Nanostructures of Paraelectric Lead Telluride

“…Let us note that it is much thinner than usually applied in other systems (e.g. heterostructures of III-V materials [3,6]). Standard transport measurements reveal the electron densities n 2D ranging from 1 × 10 12 cm −2 to 1 × 10 13 cm −2 and electron mobilities not exceeding 1 × 10 5 cm 2 /(V s) at 4.2 K. This corresponds to the electron mean free paths l e ≤ 3 µm.…”

“…Such spurious fluctuations cannot be screened out due to a very weak screening in low-dimensional systems and a low number of mobile charges. Modulation doping, in which the doped region is spatially separated from the free carriers, has been utilized as a mean to alleviate the problem, however, even in this case one still may observe strong effects caused by fluctuations produced by the long-range tails of the Coulomb potentials [6].…”