Low Temperature Anomaly in Mesoscopic Kondo Wires

Mohanty, Pritiraj; Webb, Richard A.

doi:10.1103/physrevlett.84.4481

Cited by 33 publications

(46 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…5,9,10,11,30,31 is extremely broad, ranging from 3.3 ppm, see Ref. 10, up to a few percent 30 . For concentrations n s > ∼ 100 ppm, the function ∆ρ K (T ) has a clear maximum 5,30 .…”

Section: Discussionmentioning

confidence: 94%

Electron transport and energy relaxation in dilute magnetic alloys

2003

View full text Add to dashboard Cite

We consider the effect of the RKKY interaction between magnetic impurities on the electron relaxation rates in a normal metal. The interplay between the RKKY interaction and the Kondo effect may result in a non-monotonic temperature dependence of the electron momentum relaxation rate, which determines the Drude conductivity. The electron phase relaxation rate, which determines the magnitude of the weak localization correction to the resistivity, is also a non-monotonic function of temperature. For this function, we find the dependence of the position of its maximum on the concentration of magnetic impurities. We also relate the electron energy relaxation rate to the excitation spectrum of the system of magnetic impurities. The energy relaxation determines the distribution function for the out-of-equilibrium electrons. Measurement of the electron distribution function thus may provide information about the excitations in the spin glass phase.

show abstract

“…5,9,10,11,30,31 is extremely broad, ranging from 3.3 ppm, see Ref. 10, up to a few percent 30 . For concentrations n s > ∼ 100 ppm, the function ∆ρ K (T ) has a clear maximum 5,30 .…”

Section: Discussionmentioning

confidence: 94%

Electron transport and energy relaxation in dilute magnetic alloys

2003

View full text Add to dashboard Cite

show abstract

“…This has led to an intense discussion [2 -5] as to whether quantum fluctuations can induce dephasing at T 0. While we believe that this latter scenario is theoretically excluded for electrons in a disordered metal, the presence of only a few parts per million of dynamical impurities -realized by atomic two-level systems [6] or by magnetic impurities [7][8][9][10]-may be an alternative cause of the saturation phenomenon. Indeed, it has been shown experimentally that magnetic impurities lead to an apparent saturation of 1= ' at least in some T range [9].…”

mentioning

confidence: 99%

Universal Dephasing Rate due to Diluted Kondo Impurities

et al. 2006

View full text Add to dashboard Cite

We calculate the dephasing rate due to magnetic impurities in a weakly disordered metal as measured in a weak-localization experiment. If the density n S of magnetic impurities is sufficiently low, the dephasing rate 1= ' is a universal function, 1= ' n S =fT=T K , where T K is the Kondo temperature and is the density of states. We show that inelastic vertex corrections with a typical energy transfer E are suppressed by powers of 1= ' E / n S . Therefore, the dephasing rate can be calculated from the inelastic cross section proportional to ImT ÿ jTj 2 , where T is the T matrix which is evaluated numerically exactly using the numerical renormalization group. DOI: 10.1103/PhysRevLett.96.226601 PACS numbers: 72.15.Lh, 72.15.Qm, 72.15.Rn, 75.20.Hr Dephasing, i.e., the loss of wave coherence, is a ubiquitous phenomenon in the quantum mechanics of complex systems. It is of relevance to any experiment where both interference and interactions play a role and is, therefore, of profound importance in all areas of nanoscopic and mesoscopic physics.While the basic phenomenon of dephasing as such is of a rather general nature, the concrete definition of a dephasing rate and its experimental determination vary from context to context. In this Letter, we consider the dephasing rate as determined by weak-localization (WL) measurements in metals [1]. In weakly disordered metals, the interference of electronic wave functions on time-reversed paths leads to a characteristic reduction (or enhancement in the presence of spin-orbit interactions) of the conductivity. The magnitude of this effect is controlled by the dephasing time ' -the typical time scale over which electrons get entangled with their environment (phonons, other electrons, or dynamical impurities), thereby losing the ability to interfere. Even small magnetic fields break time-reversal invariance, thus prohibiting the interference of time-reversed paths. Fitting the magnetoresistivity to WL theory is a means to determine the dephasing rate 1= ' with high precision. Surprisingly, most of these experiments [2] show a saturation of the dephasing rate at the lowest accessible temperatures T, while theoretically it is expected that in the limit T ! 0 all inelastic processes freeze out when the system approaches its (time-reversal invariant and unique) ground state. This has led to an intense discussion [2 -5] as to whether quantum fluctuations can induce dephasing at T 0. While we believe that this latter scenario is theoretically excluded for electrons in a disordered metal, the presence of only a few parts per million of dynamical impurities -realized by atomic two-level systems [6] or by magnetic impurities [7-10]-may be an alternative cause of the saturation phenomenon. Indeed, it has been shown experimentally that magnetic impurities lead to an apparent saturation of 1= ' at least in some T range [9]. In contrast, some extremely pure Au and Ag samples with a negligible concentration of impurities [10] show no saturation and seem to follow the predictions of Altshu...

show abstract

“…The problem of decoherence in weak localization was recently revisited and intensively discussed. 1,3,4 Theoretical studies by various groups [5][6][7][8][9][10] lead to a satisfactory and consistent explanation of energy relaxation experiments by Kondo impurity mediated electronelectron interaction in the gold wires by Pierre et al 1 Those gold wires were contaminated by iron impurities, and the concentration of the impurities could be independently estimated from the magnetoresistance as well as from the temperature dependence of the resistivity. As a result, it was possible to carry out a parameter free comparison of theory an experiment.…”

mentioning

confidence: 99%

Magnetic-field effects in energy relaxation mediated by Kondo impurities in mesoscopic wires

Göppert

Galperin²,

Altshuler³

et al. 2002

Phys. Rev. B

View full text Add to dashboard Cite

We study the energy distribution function of quasiparticles in voltage biased mesoscopic wires in presence of magnetic impurities and applied magnetic field. The system is described by a Boltzmann equation where the collision integral is determined by coupling to spin 1/2 impurities. We derive an effective coupling to a dissipative spin system which is valid well above Kondo temperature in equilibrium or for sufficiently smeared distribution functions in non-equilibrium. For low magnetic field an enhancement of energy relaxation is found whereas for larger magnetic fields the energy relaxation decreases again meeting qualitatively the experimental findings by Anthore et al. (condmat/0109297). This gives a strong indication that magnetic impurities are in fact responsible for the enhanced energy relaxation in copper wires. The quantitative comparison, however, shows strong deviations for energy relaxation with small energy transfer whereas the large energy transfer regime is in agreement with our findings. 72.15.Qm, 75.75.+a I. MOTIVATION AND OVERVIEWEnergy relaxation of "hot" electrons in disordered conductors at low enough energies was for a long time believed to be determined by direct interaction between electrons. Recent experiments on mesoscopic wires (see Ref. 1 and references therein) have shown that Kondo impurities (localized spins) lead to much higher energy relaxation rates than predicted by the standard theory. 2On the other hand the inelastic collisions of electrons and their spin-flips are directly related to the phase coherence time probed by weak localization effects, such as low field magnetoresistance. The problem of decoherence in weak localization was recently revisited and intensively discussed. 1,3,4Theoretical studies by various groups 5-10 lead to a satisfactory and consistent explanation of energy relaxation experiments by Kondo impurity mediated electronelectron interaction in the gold wires by Pierre et al. 1Those gold wires were contaminated by iron impurities, and the concentration of the impurities could be independently estimated from the magnetoresistance as well as from the temperature dependence of the resistivity. As a result, it was possible to carry out a parameter free comparison of theory an experiment. 7-9At the same time the copper samples in the first energy relaxation experiment by Pothier et al.11 were fitted using the concentration and Kondo temperature of the paramagnetic impurities as free parameters. The parameters obtained from energy relaxation disagree with those obtained from the magnetoresistance experiments of the same sample.12,13 Therefore, if magnetic impurities are responsible for these effects is dubious.Since the behavior of magnetic impurities is sensitive to the applied magnetic field, studies of energy relaxation in presence of the magnetic field could either rule out or validate magnetic impurities as relevant scattering process in energy relaxation. Recently, Anthore et al. 14reported results of such experiments in Cu wires that indicate a strong ...

show abstract

Low Temperature Anomaly in Mesoscopic Kondo Wires

Cited by 33 publications

References 21 publications

Electron transport and energy relaxation in dilute magnetic alloys

Electron transport and energy relaxation in dilute magnetic alloys

Universal Dephasing Rate due to Diluted Kondo Impurities

Magnetic-field effects in energy relaxation mediated by Kondo impurities in mesoscopic wires

Contact Info

Product

Resources

About