Answers are grounded in this paper and cite their sources.
We report on our measurements of resistance fluctuations as a function of magnetic field B in an Alₓ{Ga}_{1{{-}}x}As/GaAs heterostructure of etched width w=2.5 μm in the integral- and fractional-quantum-Hall-effect regimes. High-frequency fluctuations are observed near the longitudinal resistance ({R}ₓₓ$) minima for {ν}=1, 2, 3, 4, and 1/3. The quasiperiods {Δ}B({ν} = integer) of the fluctuations for integer {ν} are all {~}0.016 T, while for {ν}=1/3, the quasiperiod {Δ}B({ν} = 1/3) is {~}0.05 T, or a factor of 3 larger. The fluctuations at integer {ν} are consistent with inter-edge-state tunneling via magnetically bound states encircling a potential hill of a diameter roughly equal to the conducting width of the channel. A similar model, with the difference that the tunneling is by quasiparticles of fractional charge e*=e/q, predicts a scaling of the quasiperiod as {Δ}B({ν}=1/q)=q {Δ}B({ν} = integer).Interpreted in terms of this model, the data provide direct evidence of the existence of quasiparticles of charge e*=e/3 in the {ν}=1/3 fractional quantum Hall effect. For both {ν}=1/3 and {ν}= integer, the individual fluctuation patterns for different pairs of voltage probes are strongly correlated only if the pairs share a length of the channel, indicating that the source of the fluctuations is local, as predicted by the model. A Coulomb blockade as the origin of the fluctuations is ruled out by the fact that for {ν}=1 and 2 the fluctuation amplitudes saturate at temperatures Tc({ν}=1){}66 mK and Tc({ν}=2){}121 mK, and also saturate at currents Ic({ν}=1){}0.5 nA and Ic({ν}=2){}1.7--3.0 nA. These results indicate that for integer {ν}, the bound-state-energy spacing {Δ}{ε}({ν}) scales as {ν} or B^-1, inconsistent with a Coulomb blockade.
No takes yet. Share an insight, caveat, or question.
Simmons et al. (1991) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: