Quantum corrections to the conductivity in two-dimensional systems: Agreement between theory and experiment

On a high-mobility 2D electron gas we have observed, in strong magnetic fields (ωcτ > 1), a parabolic negative magnetoresistance caused by electron-electron interactions in the regime of kBT τ / ∼ 1, which is the transition from the diffusive to the ballistic regime. From the temperature dependence of this magnetoresistance the interaction correction to the conductivity δσ Electron-electron interaction (EEI) corrections to the Drude conductivity σ 0 of 2D systems have been intensively studied over two decades. These studies were based on the theory of interactions in the diffusive regime,. Physically this condition implies that the effective interaction time, /k B T , is larger than the momentum relaxation time τ and therefore the two interacting electrons experience scattering by many impurities. In the ballistic regime, k B T τ / > 1, electrons interact when scattered by a single impurity. A theory of the interaction correction for such a case was only recently developed [2], and there have already been several experimental attempts to apply it to the conductance of high-mobility (large τ ) semiconductor structures [3,4,5,6,7,8]. An essential feature of this theory is that the impurities are treated as point-like scatterersthe condition which is not satisfied in structures where the impurities are separated from the 2D channel by an undoped spacer (unless the spacer is thick enough for the background impurities to dominate the scattering). There is then a question of how the interaction correction in the ballistic regime manifests itself in a smooth fluctuation potential.Introducing a long-range scattering potential is expected to suppress the interaction correction in the ballistic regime considered in [2]. This correction is caused by electron back-scattering, but in the case of a smooth potential the backscattering is significantly reduced. However, as shown in [9,10], applying a strong magnetic field increases the probability of an electron to return back and restores the interaction correction.

show abstract

“…The parameter k B T τ / in our experiment is varied from 0.04 to 3.8. (This value is 3.3-33 in [12], 1.7-5.6 in [15] and 0.004-0.18 in [14]). …”

mentioning

confidence: 84%

“…However, apart from the experiment [14] on low-mobility structures, these experiments were in fact performed not in the diffusive but ballistic regime, where Eq. 1 is not justified.…”

mentioning

confidence: 99%

Magnetoresistance of a 2D Electron Gas Caused by Electron Interactions in the Transition from the Diffusive to the Ballistic Regime

Proskuryakov

Savchenko

et al. 2003

Phys. Rev. Lett.

View full text Add to dashboard Cite

show abstract

“…It should be noted that in the literature an empirical coefficient of order 2 is often introduced to bring the experimental data into better agreement with Eq. 4 10,19 . This model works well in metals, where Fermi-energy is large and the electron gas can be considered as being very uniform 4,5 , but a saturation of τ ϕ is usually reported at low temperatures (see e.g.…”

Section: Discussionmentioning

confidence: 99%

Experimental study of weak antilocalization effects in a high-mobilityInxGa1−xA
Studenikin
¹
,
Coleridge
²
,
Ahmed
³

et al. 2003
Phys. Rev. B
74
8
57
0
View full text Add to dashboard Cite

The magnetoresistance associated with quantum interference corrections in a high mobility, gated InGaAs/InP quantum well structure is studied as a function of temperature, gate voltage, and angle of the tilted magnetic field. Particular attention is paid to the experimental extraction of phasebreaking and spin-orbit scattering times when weak anti-localization effects are prominent. Compared with metals and low mobility semiconductors the characteristic magnetic field Btr =h/4eDτ in high mobility samples is very small and the experimental dependencies of the interference effects extend to fields several hundreds of times larger. Fitting experimental results under these conditions therefore requires theories valid for arbitrary magnetic field. It was found, however, that such a theory was unable to fit the experimental data without introducing an extra, empirical, scale factor of about 2. Measurements in tilted magnetic fields and as a function of temperature established that both the weak localization and the weak anti-localization effects have the same, orbital origin. Fits to the data confirmed that the width of the low field feature, whether a weak localization or a weak anti-localization peak, is determined by the phase-breaking time and also established that the universal (negative) magnetoresistance observed in the high field limit is associated with a temperature independent spin-orbit scattering time. * sergei.studenikin@nrc.ca

show abstract

“…2,3 The extraction of the EEI contribution depends on the successful determination of K ee from T-dependent conductivity measurements. However, such measurements typically also include corrections from other effects such as electron-phonon interactions and weak localization ͑WL͒.…”

Section: Introductionmentioning

confidence: 99%

Electron-electron interactions in highly disordered two-dimensional systems

2008

View full text Add to dashboard Cite

We investigate the use of three well established methods to model the electron-electron interaction correction to the conductivity in highly disordered two-dimensional systems. In order to determine the interaction correction ␦ ee , we use each method to fit experimental magnetoconductivity data from highly disordered Si:P ␦-doped two-dimensional electron systems having conductivities ranging from k F l ϳ 5 to 30. From this analysis we find that none of the three methods fits the data over the whole range of conductivities achievable.Consequently, we present a new method of extracting ␦ ee , based on the elimination of the interaction correction from the resistivity tensor, which is effective for extracting ␦ ee over the complete range of conductivities studied. We find ␦ ee between −5.5e 2 / h and −7.5e 2 / h at 0.5 K for our samples.

show abstract

Quantum corrections to the conductivity in two-dimensional systems: Agreement between theory and experiment

Cited by 59 publications

References 14 publications

Magnetoresistance of a 2D Electron Gas Caused by Electron Interactions in the Transition from the Diffusive to the Ballistic Regime

Magnetoresistance of a 2D Electron Gas Caused by Electron Interactions in the Transition from the Diffusive to the Ballistic Regime

Experimental study of weak antilocalization effects in a high-mobilityInxGa1−xA
Studenikin
¹
,
Coleridge
²
,
Ahmed
³

et al. 2003
Phys. Rev. B
74
8
57
0
View full text Add to dashboard Cite

Electron-electron interactions in highly disordered two-dimensional systems

Contact Info

Product

Resources

About

Quantum corrections to the conductivity in two-dimensional systems: Agreement between theory and experiment

Cited by 59 publications

References 14 publications

Magnetoresistance of a 2D Electron Gas Caused by Electron Interactions in the Transition from the Diffusive to the Ballistic Regime

Magnetoresistance of a 2D Electron Gas Caused by Electron Interactions in the Transition from the Diffusive to the Ballistic Regime

Experimental study of weak antilocalization effects in a high-mobilityInxGa1−xAStudenikin1, Coleridge2, Ahmed3 et al. 2003Phys. Rev. B748570View full textAdd to dashboardCite

Electron-electron interactions in highly disordered two-dimensional systems

Contact Info

Product

Resources

About

Experimental study of weak antilocalization effects in a high-mobilityInxGa1−xA
Studenikin
¹
,
Coleridge
²
,
Ahmed
³

et al. 2003
Phys. Rev. B
74
8
57
0
View full text Add to dashboard Cite