Table of Contents
Fetching ...

The role of electron interactions in a failed insulator revealed by shot noise

Mateusz Szurek, Hanqiao Cheng, Zilu Pang, Yiou Zhang, Sergei Urazhdin

Abstract

In materials known as failed insulators, electrical resistivity increases as temperature decreases, yet does not diverge - a phenomenon inconsistent with single-particle theories. We investigate the origin of this behavior by measuring shot noise in nanojunctions of nitrogen-doped beta-Ta, a prototypical failed insulator. Junctions as short as 8 nanometers exhibit hot-electron shot noise, indicating strong electron interactions. We show that charge hopping mediated by these interactions explains the anomalous electronic properties. Our findings open new avenues for exploiting electron interactions in spin-orbitronic and superconducting applications of failed insulators.

The role of electron interactions in a failed insulator revealed by shot noise

Abstract

In materials known as failed insulators, electrical resistivity increases as temperature decreases, yet does not diverge - a phenomenon inconsistent with single-particle theories. We investigate the origin of this behavior by measuring shot noise in nanojunctions of nitrogen-doped beta-Ta, a prototypical failed insulator. Junctions as short as 8 nanometers exhibit hot-electron shot noise, indicating strong electron interactions. We show that charge hopping mediated by these interactions explains the anomalous electronic properties. Our findings open new avenues for exploiting electron interactions in spin-orbitronic and superconducting applications of failed insulators.

Paper Structure

This paper contains 5 sections, 9 equations, 4 figures, 1 table.

Figures (4)

  • Figure 1: (a) Resistivity $\rho$ vs $T$ for $\alpha$-Ta, $\beta$-Ta, and TaN$_x$ deposited at the labeled $N_2$ pressure $P_{N2}$ in Torr. Inset: nitrogen concentration for the studied films estimated from the published resistivity vs $\rho(T =300\,K)$ dependence (black and brown symbols), and measured for LD-TaN$_x$ by energy dispersive spectroscopy (red symbol) [see SM for details supp]. (b) Hole density per atomic site vs $T$, for $\beta$-Ta and a TaN$_x$ sample deposited at $P_{N2}=4\times10^6$ Torr. (c) Schematic of VNJ and noise measurement setup.
  • Figure 2: Noise measurement results for the LD-Ta VNJ with $L = 8$ nm. (a) Noise vs $V_B$, at the labeled $T$. Curves: fittings with Eq. (\ref{['eq:SN_general']}). (b) $F$ vs $T$. Curve: fitting as described in the text. (c) $B$ vs $T$. Line is a guide for the eye. (d) $B$ vs $F$. The line is a linear fit with zero intercept.
  • Figure 3: (a) $F$ vs $T$ for two $\beta$-Ta nanowires with the same length $L=500$ nm and cross-section areas, but different contact area with the substrate as determined by thickness. Curves: fittings as described in the text. (b), (c) Fitting parameters in the dependence $f_0=a+bT^3$ vs $L$, for VNJs and nanowires as labeled. Lines are guides for the eye. To maximize the precision of $a$, fitting was performed for $T\leq10$ K. Inset in (b): nanowire schematic. (d) $F$ vs $L$ at $T = 4.2$ K for LD-Ta VNJs ($L\leq20$ nm) and $\beta$-Ta nanowires ($L\geq 100$ nm).
  • Figure 4: (a) $\rho$ vs $T$ for HD-Ta and best fits with Mott and ES VRH models, as labeled. (b) Spectral density functions of two quasi-localized states for negligible lifetime broadening ($\gamma = 0$) and finite lifetime due to e-e interaction ($\gamma>0$). (c) $\rho$ vs $T$ for $\beta$-Ta and a series of nitrogen-doped Ta films, with fittings using the model of hopping mediated by the e-e interaction, as discussed in the text. (d) $\sigma_{e-e}$ and $\sigma_{e-ph}$ vs nitrogen pressure at $T = 300$ K.