Noise-to-current ratio divergence as a fingerprint of dispersing Majorana edge modes
Leo Katayama, Andreas P. Schnyder, Yasuhiro Asano, Satoshi Ikegaya · 2025
Read Noise-to-current ratio divergence as a fingerprint of dispersing Majorana edge modes by Leo Katayama, Andreas P. Schnyder, Yasuhiro Asano, Satoshi Ikegaya (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-20921.txt.
Introduction
Majorana edge modes are among the most wanted particles in physics, and every claimed sighting so far has an alibi. Leo Katayama and co-authors propose a test they call a smoking gun in a 2025 paper. The device joins two normal metal leads to a topological superconductor carrying the dispersing modes.
Because the modes are unpaired, they carry fluctuation without carrying average charge current in the linear regime. The noise-to-current ratio therefore blows up, a divergence no ordinary mode should fake. The claim is proven analytically, without leaning on any one material model. Tight-binding simulations of two real candidate platforms confirm the divergence numerically. The authors discuss how the test could run on existing heterostructures.
Catalog record
| CREATOR | Leo Katayama, Andreas P. Schnyder, Yasuhiro Asano, Satoshi Ikegaya |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 46 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-20921
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Noise-to-current ratio divergence as a fingerprint of dispersing Majorana edge modes Preprint: APS/123-QED Leo Katayama, Andreas P. Schnyder, Yasuhiro Asano, Satoshi Ikegaya Abstract The definitive detection of Majorana modes in topological superconductors is a key issue in condensed matter physics. Here we propose a smoking-gun experiment for the detection of one-dimensional dispersing Majorana edge modes, based on theoretical results for multi-terminal transport in a setup consisting of two normal metal leads and a topological superconductor. In the proposed device, the unpaired nature of the Majorana edge modes inherently leads to the absence of the charge current in the linear response regime, while the current fluctuation remains significant. Therefore, the divergence in the noise-to-current ratio serves as unambiguous evidence for the presence of the dispersing Majorana edge modes. We reach this conclusion analytically, without relying on any specific model of topological superconductors. In addition, using tight-binding models of topological-insulator-based topological superconductors, we numerically verify the predicted divergent noise-to-current ratio. We also discuss the application of our proposal to the CoSi 2 /TiSi 2 heterostructure and the iron-based superconductor FeTe 1-x Se x . I Introduction Majorana modes in topological superconductors (TSs) green_00 ; kitaev_01 ; wilczek_09 ; kane_10 ; zhang_11 have become a central topic in condensed matter physics, due to their potential applications for fault-tolerant topological quantum information technologies. ivanov_01 ; kitaev_03 ; sarma_08 ; fisher_11 ; alicea_16 . To date, the existence of Majorana edge modes (MEMs) has been predicted in various platforms, including semiconductor-superconductor hybrids sato_09(1) ; sarma_10(1) ; sarma_10(2) ; oreg_10 ; kouwenhoven_12 ; flensberg_17 ; halperin_17 ; nichele_19 ; yacoby_19 , magnetic material-superconductor hybrids beenakker_11 ; yazdani_13 ; yazdani_14 ; cren_17 , topological insulator-superconductor hybrids kane_08 ; zhang_10 ; zhang_15 , Cu x Bi 2 Se 3 ando_11 ; fu_12 ; tanaka_12 , and iron-based superconductors kane_13 ; zfang_17 ; zhang_16 ; pan_15 ; shin_18 ; gao_18 ; tamegai_19 ; sarma_21 . However, the unambiguous verification of MEMs in these systems remains a significant challenge microsoft_25 . Zero-bias conductance peaks observed in normal metal–superconductor junctions buchholtz_81 ; hu_94 ; tanaka_95 ; sarma_01 ; asano_04 ; law_09 have long been regarded as a strong experimental signature of MEMs. Nevertheless, recent studies have shown that zero-bias conductance peaks can also arise from various factors unrelated to Majorana physics brouwer_12 ; aguado_12 ; sarma_17 ; klinovaja_20 ; chang_20 ; katsaros_21 ; frolov_21 ; ando_24 , making their definitive identification difficult. This stalemate highlights the necessity of further experimental strategies for the unambiguous detection of MEMs. Figure 1: Schematic image of the setup consisting of two NM leads and a TS. We measure the charge current I_{\alpha} in the \alpha -th NM due to the bias voltage V_{\beta} applied to the \beta -th NM, where \alpha\neq\beta . The \alpha -th NM and the TS are grounded (i.e., V_{\alpha}=0 ). In Figure, we illustrate the case for (\alpha,\beta)=(2,1) , where a dispersing MEM mediates inter-lead scattering from an electron in the first NM to either an electron or a hole in the second NM. In this Letter, we propose a versatile experiment that provides a smoking-gun signature of the one-dimensional dispersing MEMs. The one-dimensional dispersing MEMs are particularly important because they can be harnessed in the braiding operations of topological computing beenakker_19 ; beenakker_19(2) ; beenakker_20 . To detect the dispersing MEMs, we measure the nonlocal charge transport law_09 ; akhmerov_09 ; kane_09(2) ; beenakker_10 ; beri_12 ; beenakker_15(2) ; ikegaya_19 in a multi-terminal setup as shown in Fig. 1 , where two normal metal (NM) leads are attached to one edge of a TS. When the distance between the NM leads ( L ) is sufficiently larger than the superconducting coherence length ( \xi ), inter-lead scattering mediated by evanescent modes of conventional Bogoliubov quasiparticles vanishes completely deutscher_00 . Thus, the nonlocal charge current at bias voltages below the superconducting gap is carried exclusively by the MEM propagating along the edge. Importantly, in usual Bogoliubov quasiparticles, electron-like and hole-like modes exist in pairs at zero energy due to particle-hole symmetry. In contrast, the MEM, where particle and antiparticle modes are equivalent, can exist individually. We show that this unpaired nature of the MEM essentially leads to the absence of the charge currents in the linear response limit, while the current fluctuation remains significant. As a result, the divergence in the noise-to-current ratio at zero bias voltage serves as a smoking-gun signature of the dispersing MEM in the proposed setup. We reach this conclusion without relying on any specific model of the TS. Thus, we expect that the proposed experiment is useful to detect any one-dimensional dispersing MEMs, including Majorana hinge modes in three-dimensional second-order TSs balents_15 ; hughes_17 ; fang_17 ; brouwer_17 ; khalaf_18 . To validate our theory more robustly, we also perform numerical calculations for specific tight-binding models of the superconducting topological insulator thin films zhang_15 ; sarma_21 . Finally, we discuss potential applications of our proposal to the CoSi 2 /TiSi 2 heterostructure lin_21 ; lin_23 and the iron-based superconductor FeTe 1-x Se x sarma_19 ; kim_21 ; zhang_24 ; burch_19 . II noise-to-current ratio Let us consider a two-dimensional hybrid system as shown in Fig. 1 , where two NM leads are attached to an edge of a TS. We calculate the charge current at zero temperature within the Blonder–Tinkham–Klapwijk (BTK) formalism deutscher_00 ; klapwijk_82 : \displaystyle\begin{split}&\bar{I}_{\alpha}=\frac{1}{e}\int^{eV_{\beta}}_{0}G_{\alpha\beta}(E)dE,\\ &G_{\alpha\beta}(E)=-\frac{e^{2}}{h}\mathrm{Tr}\left[\hat{T}^{e}_{\alpha\beta}(E)-\hat{T}^{h}_{\alpha\beta}(E)\right],\\ &\hat{T}^{\gamma}_{\alpha\beta}(E)=\{\hat{s}^{\gamma e}_{\alpha\beta}(E)\}^{\dagger}\hat{s}^{\gamma e}_{\alpha\beta}(E),\end{split} (1) where we assume \alpha\neq\beta . \bar{I}_{\alpha} represents the time-averaged charge current in the \alpha -th NM due to the bias voltage V_{\beta} applied to the electrode connected to the \beta -th NM, where the electrodes attached to the \alpha -th NM and the TS are grounded (i.e., V_{\alpha}=0 ). We denote the number of propagating channels in the \alpha -th NM by N_{\alpha} . The N_{\alpha}\times N_{\beta} matrices of \hat{s}^{ee}_{\alpha\beta} and \hat{s}^{he}_{\alpha\beta} ( \hat{s}^{eh}_{\alpha\beta} and \hat{s}^{hh}_{\alpha\beta} ) contain the scattering coefficients from an electron (a hole) in the \beta -th NM to an electron and a hole in the \alpha -th NM at energy E , respectively. In addition, we calculate the zero-frequency noise power based on the BTK formalism beenakker_94 ; datta_96 , \displaystyle C_{\alpha}=\int^{\infty}_{-\infty}\overline{\delta I_{\alpha}(0)\delta I_{\alpha}(\tau)}d\tau=\frac{1}{e}\int^{eV_{\beta}}_{0}P_{\alpha\beta}(E)dE, (2) with \displaystyle\begin{split}P_{\beta\alpha}(E)=&\frac{e^{3}}{h}\mathrm{Tr}\left[\hat{T}^{e}_{\alpha\beta}(E)+\hat{T}^{h}_{\alpha\beta}(E)\right]\\ -&\frac{e^{3}}{h}\mathrm{Tr}\left[\{\hat{T}^{e}_{\alpha\beta}(E)-\hat{T}^{h}_{\alpha\beta}(E)\}^{2}\right],\end{split} (3) where \delta I_{\alpha}(\tau)=I_{\alpha}(\tau)-\bar{I}_{\alpha} represents the deviation of the current at time \tau from the time-averaged current. In the BTK formalism, we assume that the currents flowing towards x=+\infty ( x=-\infty ) are absorbed into the ideal electrode connected to the TS (NM), while these electrodes are not explicitly included in the Hamiltonian. The BTK formalism is quantitatively valid for bias voltages well below the superconducting gap. In the following, we focus on the linear response limit V_{\beta}\rightarrow 0 , where the time-averaged current, \bar{I}_{\alpha}=G_{\alpha\beta}(0)V_{\beta} , and the zero-frequency noise power, C_{\alpha}=P_{\alpha\beta}(0)V_{\beta} , are calculated using only the inter-lead scattering matrix at zero energy: \displaystyle\hat{s}_{\alpha\beta}(E=0)=\left[\begin{array}[]{cc}\hat{s}^{ee}_{\alpha\beta}&\hat{s}^{eh}_{\alpha\beta}\\ \hat{s}^{he}_{\alpha\beta}&\hat{s}^{hh}_{\alpha\beta}\end{array}\right],\qquad(\alpha\neq\beta), where we use the notation \hat{s}^{\gamma\gamma^{\prime}}_{\alpha\beta}(E=0)=\hat{s}^{\gamma\gamma^{\prime}}_{\alpha\beta} , for simplicity. Due to particle-hole symmetry of the superconducting system, the scattering matrix satisfies beenakker_15 \displaystyle\begin{split}\hat{\Xi}_{\alpha}\hat{s}_{\alpha\beta}(E)\hat{\Xi}_{\beta}=\hat{s}_{\alpha\beta}(-E),\quad\hat{\Xi}_{\alpha}=\left[\begin{array}[]{cc}\hat{O}_{\alpha\alpha}&\hat{I}_{\alpha}\\ \hat{I}_{\alpha}&\hat{O}_{\alpha\alpha}\end{array}\right]\mathcal{K},\end{split} (6) where \hat{O}_{\alpha\beta} is the N_{\alpha}\times N_{\beta} null matrix, \hat{I}_{\alpha} denotes the N_{\alpha}\times N_{\alpha} identity matrix, and \mathcal{K} represents the complex conjugation operator. Therefore, we obtain \displaystyle\hat{s}^{eh}_{\alpha\beta}=(\hat{s}^{he}_{\alpha\beta})^{\ast},\quad\hat{s}^{hh}_{\alpha\beta}=(\hat{s}^{ee}_{\alpha\beta})^{\ast}. (7) We rewrite the scattering matrix in a Majorana representation, transforming \hat{s}_{\alpha\beta}(E=0) into a real