Superconducting Diode Effect in Weak Localization Regime

Naratip Nunchot, Youichi Yanase · 2025

Read Superconducting Diode Effect in Weak Localization Regime by Naratip Nunchot, Youichi Yanase (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-2507-21897.txt.

Introduction

Naratip Nunchot and Youichi Yanase wrote a paper from 2025 on a superconductor that lets more current pass one way than the other. Such a one-way effect could serve in future circuits. The pair studied a dirty, flat superconductor with two added traits: a twist that links electron spin to motion, and a magnetic field lying in the plane. Dirt and weak trapping of electrons make the theory hard to get right.

They built a full theory that counts the push between electrons and the weak trapping effect at a first level of detail. Heat and field limits for the super state both shrank once those forces were counted. The one-way effect, though, held firm in the hot range they studied. They also worked out how the material carries current once the super state breaks. There, the trapping effect showed up clearly as the current slowed and stuck.

Catalog record

CREATORNaratip Nunchot, Youichi Yanase
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE27 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-21897

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Superconducting Diode Effect in Weak Localization Regime Naratip Nunchot, Youichi Yanase Abstract We study a dirty two-dimensional superconductor with Rashba spin-orbit coupling and in-plane Zeeman fields described by the nonlinear sigma model that includes short-range electron-electron interactions from the Coulomb and Cooper channels. The renormalized Ginzburg-Landau theory, which includes the weak localization effects at the one-loop level, is constructed by using the Keldysh functional formalism. It is shown that the tricritical point appears in the phase diagram. The superconducting diode quality factor increases divergently as the system approaches the tricritical point. Near the superconducting phase transition lines, the absolute value of the diode quality factor decreases due to the cooperation of localization and interactions. The normal conductivity of the resistive state, in which the superconducting state is destroyed by the critical current, is calculated, and localization behaviors are demonstrated. Introduction: The phenomenon in which the critical currents of superconductors become nonreciprocal, known as the superconducting diode (SD) effect Ando ; Bauriedl , has attracted much attention. Theoretical studies have been carried out for various types of noncentrosymmetric superconductors Yuan ; Daido1 ; He-Nagaosa ; Aoyama ; Nunchot1 ; Banerjee ; Nunchot2 . Not only clean systems but also systems with uncorrelated impurities have been studied in detail at the mean-field level Ilic1 . If electron systems are disordered enough, they are well described by the nonlinear sigma model (NLSM) Wegner ; Hikami ; Finkelstein ; Kamenev1 , and dirty superconductors can also be described by the NLSM Feigelman ; Yurkevich ; Levchenko1 ; Konig . Recently, an NLSM for a dirty superconductor with spin-orbit coupling (SOC) and a Zeeman field was derived Virtanen and used to investigate the SD effect in Josephson junctions at the mean-field level Ilic2 . However, the Anderson localization effect , a correction to the mean-field theory in dirty systems, has not yet been highlighted in studies of the SD effect. The role of Anderson localization in superconductors is an important topic of ongoing debate. Without considering strong mesoscopic correlations and electron-electron (e-e) interactions, the properties of dirty s -wave superconductors at zero magnetic field remain unchanged with respect to the strength of non-magnetic disorder in the weak localization (WL) regime, known as the Anderson theorem Anderson . If the Coulomb interaction is concerned, the transition temperature for the s -wave pairing states is suppressed in the WL regime Maekawa ; Finkelstein1 ; Oreg . The opposite effect can occur in the strong localization regime by the multifractality effect Feigelmanloffe ; Burmistrov1 ; Burmistrov11 ; Burmistrov2 . It was also shown that, in the WL regime, the Coulomb interaction suppresses the critical current of conventional dirty superconductors at zero magnetic field and low temperatures Fukuyama ; Takayanagi . In this Letter, we study the interplay of the e-e interaction and disorder on the SD effect and clarify the WL correction to conductivity when the superconducting (SC) state is broken by the critical current. We work with the NLSM for a dirty two-dimensional (2D) noncentrosymmetric superconductor, including the interactions from the Coulomb and Cooper channels in the short-range limit, and establish a renormalized Ginzburg-Landau (GL) theory at the one-loop level. While many studies Burmistrov2 ; Andriyakhina1 ; Nosov ; Andriyakhina2 have been conducted in the replica formalism, we formulate the theory with the Keldysh functional formalism Kamenev1 ; Feigelman ; Levchenko1 ; Schwiete1 . Model: We consider a dirty 2D superconductor aligned in the xy -plane with the Rashba SOC and a Zeeman field in the x -direction. In the Keldysh functional formalism, the partition function is given as Z=\int D[Q,\Delta]\exp\left({iS[Q,\Delta]}\right) with Virtanen \displaystyle S[\Delta,Q] \displaystyle=S_{\rm HS}[\Delta]+S_{\rm N}[\Delta,Q]+S_{\rm T}[Q], (1) where S_{\rm HS} is responsible for the Hubbard-Stratonovich transformation associated with the SC state, S_{\rm N} is the conventional non-interacting part, and S_{\rm T} is the topological part. Their explicit forms are shown in Supplemental material Sup . A 8\times 8 matrix field Q satisfies the conditions Q^{2}=1 and \mathrm{Tr}\,\{Q\}=0 . Let \gamma_{\mu},~\tau_{\mu} , and \sigma_{\mu}~(\mu=0,\dots,4) be the Pauli matrices in the Keldysh, Nambu, and spin spaces, respectively. In particular, we write \gamma_{\rm c}=\gamma_{0} and \gamma_{\rm q}=\gamma_{1} and often abbreviate \gamma_{0} for simplicity. The order parameter of superconductivity, that is, the SC gap, is then described by the matrix \gamma_{\rm c}\check{\Delta}_{\rm c}+\gamma_{\rm q}\check{\Delta}_{\rm q} , where \check{\Delta}_{\rm c(q)}=\tau_{+}\Delta_{\rm c(q)}+\tau_{-}\Delta^{*}_{\rm c(q)} with \tau_{\pm}=(\tau_{1}\pm i\tau_{2})/2 and \Delta_{\rm c(q)} is the classical (quantum) component of the SC gap. The check symbol \check{(\cdot)} highlights that the objects are 4\times 4 matrices. The terms in the form \gamma_{\mu}\check{A} stand for the expression \gamma_{\mu}\otimes\check{A} , where \otimes is a tensor product. We assume that the SC state is a helical state modulating in the y -direction and time-independent. In this case, the Cooper pair momentum corresponds to a uniform gauge field along the y -axis A_{y} , and we set A_{x}=0 from now on. In addition to Eq. ( 1 ), we take into account interaction terms of the Cooper and Coulomb channels. The former can be obtained directly by integrating the spatial fluctuation of the gap function from Eq. ( 1 ) Konig ; Burmistrov2 . The procedure results in two terms in action. The first term \bar{S}_{\rm HS} has the same form as S_{\rm HS} [Eq. (S.1) in Supplemental Material Sup ], but the spatial modulation in the gap functions is removed. Thus, in the following, \Delta_{\rm c(q)} denotes the spatially uniform gap function. The second term is \displaystyle S^{\rm(c)}_{\rm int}[Q]=\tilde{\Gamma}_{\rm c}\!\!\sum\limits_{i=1,2}\!\int_{t,\bm{r}}\!\!\!\mathrm{tr}\!\left\{{\gamma_{\rm c}\tau_{i}\delta Q_{tt}(\bm{r})}\right\}\!\mathrm{tr}\!\left\{{\gamma_{\rm q}\tau_{i}\delta Q_{tt}(\bm{r})}\right\}\!, (2) where \int_{t,\bm{r}}\equiv\int dtd\bm{r} , \tilde{\Gamma}_{\rm c}=-\pi^{2}\nu\Gamma_{\rm c}/16 , \nu=m/(2\pi) , m is the electron mass, \Gamma_{\rm c} is the SC coupling constant, and the trace \mathrm{tr}\,\{\cdot\} is performed only in the matrix elements. The spatial fluctuation is represented by \delta Q=Q-\bar{Q} , where \bar{Q} denotes the spatial average of Q . For the Coulomb channel in the short-range limit, following the procedure in Ref. Schwiete1 , we came to the action as follows: \displaystyle S^{\rm(s)}_{\rm int}[Q]=\tilde{\Gamma}_{\rm s}\int_{t,\bm{r}}\mathrm{tr}\left\{{\gamma_{\rm c}\tau_{0}Q_{tt}(\bm{r})}\right\}\mathrm{tr}\left\{{\gamma_{\rm q}\tau_{0}Q_{tt}(\bm{r})}\right\}, (3) where \tilde{\Gamma}_{\rm s}=-\pi^{2}\nu\Gamma_{\rm s}/16 and \Gamma_{\rm s}=1 in the short-range limit. Thus, the total action in the partition function Z is given by S_{\rm tot}=\bar{S}_{\rm HS}+S_{\rm N}+S_{\rm T}+S^{\rm(c)}_{\rm int}+S^{\rm(s)}_{\rm int} . Effective action: In the ordinary mean-field theory, the system is described by saddle points of action, which are obtained by searching the matrix field Q=\bar{Q} that optimizes the action under the constraints \bar{Q}^{2}=1 and \mathrm{Tr}\,\{\bar{Q}\}=0 . Going beyond the mean-field approximation can be done by integrating out the fluctuations around the saddle points, resulting in the effective action of the gap function. In this work, we consider the Gaussian fluctuations around a SC saddle point \bar{Q}_{\rm s} . Since \bar{Q}_{\rm s} must be on the same manifold as a normal saddle point \bar{Q}_{\rm N} , there is a matrix U such that \bar{Q}_{\rm s}=U\bar{Q}_{\rm N}U^{-1} . The saddle point configuration may be taken as the fermionic action formulated by Keldysh path integration Kamenev1 . That is \bar{Q}=U_{\rm K}\Lambda U_{\rm K} , where \Lambda=\mathrm{diag}\,\{\check{\Lambda}_{\rm R},\check{\Lambda}_{\rm A}\} and \rm R(A) denotes the retarded (advanced) component. In the energy representation, U_{\rm K} is \displaystyle U_{\mathrm{K}\epsilon} \displaystyle=\begin{pmatrix}\tau_{0}&F_{\epsilon}\tau_{0}\\ 0&-\tau_{0}\end{pmatrix}, (4) where distribution in thermal equilibrium leads to F_{\epsilon}=\tanh\epsilon/(2T) with energy \epsilon and temperature T . Using the above equations, we can rewrite \bar{Q}_{\rm N(s)}=U_{\rm K}\Lambda_{\rm N(s)}U_{\rm K} ,

Read the complete text