Table of Contents
Fetching ...

Limits of Thermal Conductance Quantization in Chiral Topological Josephson Junctions

Daniel Gresta, Fernando Dominguez, Raffael L. Klees, Florian Goth, Laurens W. Molenkamp, Ewelina M. Hankiewicz

TL;DR

This work addresses how chiral Majorana modes in multiterminal topological Josephson junctions manifest in thermal versus nonlocal electrical transport. By combining a lattice-regularized Dirac-BdG model with non-equilibrium Green's function transport, it identifies clear conditions under which a single chiral Majorana channel yields half-quantized thermal conductance $\kappa=0.5\kappa_0$ at phase difference $\phi=\pi$, while nonlocal conductance remains suppressed due to particle-hole symmetry. The study shows that in the $C=1$ phase, half-quantized thermal plateaus persist in intermediate-to-long junctions and at finite Zeeman fields, but in the $C=2$ phase the thermal response is not universally quantized and strongly depends on the momentum-space location of Majorana modes. The results emphasize that heat transport signatures hinge on momentum-space structure, finite-size geometry, and sample parameters, offering practical criteria to identify Majorana physics in multiterminal topological superconductors and guiding extensions to other platforms.

Abstract

We investigate thermal and non-local electrical transport in four-terminal Josephson junctions formed by a normal region coupled to two transverse chiral superconducting leads, supporting phases characterized by Chern numbers ${\cal C}=0,\,1$\,and\,2. We identify the conditions under which a single chiral Majorana mode (${\cal C}=1$) produces a robust half-quantized thermal conductance, while non-local electrical conductance remains strongly suppressed by particle-hole symmetry. Thermal conductance quantization occurs near a superconducting phase difference $π$, but only in the low-doping regime of the central region and in the intermediate- to long-junction limits. At finite Zeeman fields, the thermal response broadly follows the topology of the isolated superconducting leads for the $C=1$ phase while, in the ${\cal C}=2$ phase, the thermal conductance generally deviates from quantization, depending on the momentum-space location of the Majorana modes. Our results establish clear criteria for probing chiral Majorana modes in Josephson junctions and highlight the essential role of momentum-space structure, finite-size geometry, and sample parameters in thermal transport.

Limits of Thermal Conductance Quantization in Chiral Topological Josephson Junctions

TL;DR

This work addresses how chiral Majorana modes in multiterminal topological Josephson junctions manifest in thermal versus nonlocal electrical transport. By combining a lattice-regularized Dirac-BdG model with non-equilibrium Green's function transport, it identifies clear conditions under which a single chiral Majorana channel yields half-quantized thermal conductance at phase difference , while nonlocal conductance remains suppressed due to particle-hole symmetry. The study shows that in the phase, half-quantized thermal plateaus persist in intermediate-to-long junctions and at finite Zeeman fields, but in the phase the thermal response is not universally quantized and strongly depends on the momentum-space location of Majorana modes. The results emphasize that heat transport signatures hinge on momentum-space structure, finite-size geometry, and sample parameters, offering practical criteria to identify Majorana physics in multiterminal topological superconductors and guiding extensions to other platforms.

Abstract

We investigate thermal and non-local electrical transport in four-terminal Josephson junctions formed by a normal region coupled to two transverse chiral superconducting leads, supporting phases characterized by Chern numbers \,and\,2. We identify the conditions under which a single chiral Majorana mode () produces a robust half-quantized thermal conductance, while non-local electrical conductance remains strongly suppressed by particle-hole symmetry. Thermal conductance quantization occurs near a superconducting phase difference , but only in the low-doping regime of the central region and in the intermediate- to long-junction limits. At finite Zeeman fields, the thermal response broadly follows the topology of the isolated superconducting leads for the phase while, in the phase, the thermal conductance generally deviates from quantization, depending on the momentum-space location of the Majorana modes. Our results establish clear criteria for probing chiral Majorana modes in Josephson junctions and highlight the essential role of momentum-space structure, finite-size geometry, and sample parameters in thermal transport.
Paper Structure (22 sections, 11 equations, 12 figures)

This paper contains 22 sections, 11 equations, 12 figures.

Figures (12)

  • Figure 1: Schematic of the four-terminal Josephson junction. A normal region of size $n_x \times n_y$ and chemical potential $\mu_c$ is coupled to two normal leads (left $L$ and right $R$) and two superconducting leads (top $T$ and bottom $B$). The superconducting terminals are grounded, share a common chemical potential $\mu_s$, and are kept at temperature $T_s$, with pairing potentials $\Delta e^{i\varphi_{T,B}}$. The right normal lead is also grounded ($V_R=0$), while the left normal lead is biased by a voltage $V_L$ and a temperature difference $T_L$, providing the sole source of nonequilibrium. Black bars indicate grounded terminals.
  • Figure 2: Bulk energy spectrum of the normal ($\Delta=0$) Hamiltonian, with $\mathcal{Z}=0$, $\mu=0$ and two different values of the Wilson mass, $m_0=0$ (a) and $m_0=0.5$ (b). Energy spectrum for a nanoribbon with $n_y=20a$, also for $m_0=0$ (c) and $m_0=0.5$ (d). All panels are in units of $t$.
  • Figure 3: Top panel: Phase diagram of the bulk Hamiltonian as a function of $\mu_s$ and $\mathcal{Z}$, with $m_0=0.8$ and $\Delta=0.35$, delimited by blue curves marking gap closings at high symmetry points. Different colors, white, yellow and blue set $\mathcal{C}=0,\,1,\,2$, respectively. Panels (a)-(h) show exemplary energy spectra of a nanoribbon, with a width of $n_x=60a$. (a) $(\mathcal{Z}, \mu_s)$, (a) $(0.5,0.0)$, (b) $(0.0,0.0)$, (c) $(-0.5,1.5)$, (d) $(-0.75,0.2)$, (e) $(-1.7,0.2)$, (f) $(-2.5,0.2)$, (g) $(-1.5,2.0)$ and (h) $(-2.0,2.0)$.
  • Figure 4: Low-doping regime ($\mu_\text{c}=0$): Phase-dependent DOS for a JJ with ${\cal Z}=0$, $\mu_\text{c} = 0$, $\Delta=0.8$, $\mu_s=0.8$ and (a, c) $n_y=2a$ and (b, d) $n_y=20a$ and PBC (a, b) and HWBC (c, d).
  • Figure 5: Finite-doping regime ($\mu_\text{c}=0.8$ and $n_y=20a$): (a, b) PBC as a function of $E$ and $k_x$, with (a) $\mu_\text{c}=0$ and (b) $\mu_\text{c}=0.8$ and $\phi=\pi$. (c, d) HWBC DOS as a function of energy and $\phi$ for the intermediate-junction regime, with $\mu_s=0.8$ and $\Delta=0.35$. The rest of the parameters are the same as in Fig. \ref{['fig:ABS-LDOS']}. (d) LDOS as a function of $x$ and $y$ for $E=0$ and $\phi=\pi$.
  • ...and 7 more figures