Table of Contents
Fetching ...

Unveiling local magnetic moments in copper-oxide atomic junctions

Marcel Strohmeier, Samanwita Biswas, Wolfgang Belzig, Regina Hoffmann-Vogel, Elke Scheer

Abstract

Incorporating oxygen into metallic atomic-scale junctions modifies the interatomic bonding and may even promote the formation of monoatomic chains. In the specific case of copper oxide, first-principles studies have predicted the emergence of ferromagnetic ground states, attributing certain atomic configurations with spin filtering capabilities. By means of low-temperature transport measurements, we provide a series of experimental evidence indicating the presence of local magnetism in air oxidized mechanically controllable copper break junctions. Our investigations on ultimately small contacts range from magnetotransport measurements to the analysis of anomalous shot noise in the presence of strong zero-bias anomalies. The analysis of the latter allows to determine the spin polarization (SP) of the current and that is interpreted with the Kondo physics picture.

Unveiling local magnetic moments in copper-oxide atomic junctions

Abstract

Incorporating oxygen into metallic atomic-scale junctions modifies the interatomic bonding and may even promote the formation of monoatomic chains. In the specific case of copper oxide, first-principles studies have predicted the emergence of ferromagnetic ground states, attributing certain atomic configurations with spin filtering capabilities. By means of low-temperature transport measurements, we provide a series of experimental evidence indicating the presence of local magnetism in air oxidized mechanically controllable copper break junctions. Our investigations on ultimately small contacts range from magnetotransport measurements to the analysis of anomalous shot noise in the presence of strong zero-bias anomalies. The analysis of the latter allows to determine the spin polarization (SP) of the current and that is interpreted with the Kondo physics picture.
Paper Structure (15 sections, 30 equations, 9 figures)

This paper contains 15 sections, 30 equations, 9 figures.

Figures (9)

  • Figure 1: Analysis of the opening traces recorded on an air oxidized Cu MCBJ, comparing measurements taken during two consecutive cool-downs. (a) Cartoon of the oxidized atomic contact. The light gray areas indicate the metallic Cu, the dark gray the oxidized areas at the surface of the leads and in the contact region. (b) Conductance histogram constructed from all opening curves. (b) Distribution of chain lengths extracted from the traces. (c) Examples of individual opening traces. Gray/red: First cool-down (without/with breaking the junction at room temperature)
  • Figure 2: Magnetoconductance (MC) measurements on atomic Cu/Cu$_x$O contacts. (a) Representative non-monotonic MC traces recorded after ambient-air exposure in a semi-broken state. Curves are scaled and vertically off-set for clarity. (b) Corresponding $I$–$V$ characteristics measured prior to each MC sweep, shown with vertical offsets only (see conductance scale bar). In all measurements, the magnetic field is applied perpendicular to the sample plane and swept between $\pm5$ T or $\pm6$ T. Blue (red) traces denote the return (forward) sweep. Labels indicate the zero-bias conductance (bold) and the maximum magnetoconductance ratio MCR$_\text{max}$ (in brackets). (c) MCR$_\text{max}$ as a function of zero-field conductance for all continuous MC sweeps of the same sample; different symbols distinguish sweep directions. (d) Example MC trace measured during a second cool-down after air exposure, showing nonlinear and hysteretic behavior at $G<1\,G_0$ (upper panel). The simultaneously recorded sample temperature remains constant throughout the field sweep (lower panel).
  • Figure 3: Analysis of the Kondo effect in an oxidized Cu MCBJ: (a), (b) Examples of d$I$/d$V$ curves exhibiting Fano line shapes. A fit to Eq. \ref{['eq:fano_line_shape']} is shown by the solid line. (c) Histogram of the Kondo temperature $T_K$ and occupation number $n_d$ (inset) across all contacts with $A>0.01\,G_0$. A log-normal distribution is fitted to the $T_K$ data. (d) Scatter plot of $A$ versus $T_K$. (e) Polar plot with $\alpha = \arctan(q)$, representing the distribution of the shape factor $|q| \in [0,\,\infty)$.
  • Figure 4: (a) Circuit diagram of the electronic setup for measuring shot noise at low temperatures. (b) A Cartoon depicting the spin selectivity of an atomic Cu$_x$O junction. (c) Fano factor analysis in the $F(G)$ space, with the red data points highlighting contacts entering the spin polarized transport regime (red area). The noise spectra for two contact with SP along with fits to Eq. \ref{['eq:1/f_fit']} are shown in (d). Black spectra correspond to the zero-bias (thermal) noise, while gray spectra are obtained under non-equilibrium conditions with increasing bias from bottom to top and shifted vertically for clarity. The plots (e) display the noise-voltage characteristics in reduced units $X$ and $Y$, together with a linear fit to the function $Y = F(X - 1)$ to extract the Fano factor $F$. The d$I$/d$V$ curves are shown in the insets.
  • Figure 5: Noise modeling on an atomic contact exhibiting a strong Kondo resonance: (a) $IV$ and d$I$/d$V$ characteristics, together with the Fano fit used to extract $T(E)$. (b) Excess noise $S_{I,\text{ex}}(V)$ compared with noise modeling that includes an energy dependent transmission. Calculations for the spin degenerate single-channel model (dash-dotted line) and the 2CM (red line) are shown. (c) Sketch of the 2CM applied to a Cu$_x$O atomic junction: A spin degenerate channel with a bias independent transmission $\tau_0$, together with a spin polarized channel, defines the transmission of the monoatomic constriction. The polarized channel exhibits an energy dependent transmission function $\tau_1(E)$ and carries the profile of the Kondo resonance, which is induced by the magnetic moments of the Cu-O units. Additional spin filtering may occur in the nanoscale vicinity of the apex atoms, where surface magnetism in oxidized Cu becomes more dominant. (d) Transmission functions underlying the 2CM in (b) with $SP=100\%$ and $R=3.13$ (e) Conductance versus displacement before and after the noise measurement. (f) $F(G)$ plot interpreted within the 2CM. Two exemplary lines with fixed values of $R$ and $SP$ are indicated by colored marker symbols.
  • ...and 4 more figures