Uniform response theory of non-Hermitian systems: Non-Hermitian physics beyond the exceptional point

Subhajyoti Bid, Henning Schomerus · 2024

Read Uniform response theory of non-Hermitian systems: Non-Hermitian physics beyond the exceptional point by Subhajyoti Bid, Henning Schomerus (2024) 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-2412-11932.txt.

Introduction

Subhajyoti Bid and Henning Schomerus build one response theory for non Hermitian systems that works across all spectral cases, in a 2024 paper. Such systems show striking response effects, above all near exceptional points where the state set collapses. Past theory treated each spectral case apart, which tangles the picture near the very points where response changes kind.

Their uniform expansions rest only on values computed straight from the system operator, and those values vary smoothly as a singular point nears. Two uses follow: exact conditions for higher order degeneracies, and a ladder of response strengths that runs without breaks across parameter space. Tests on an active microcavity with feedback show the frame holds in practice.

Catalog record

CREATORSubhajyoti Bid, Henning Schomerus
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE118 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2412-11932

Browse the full library · About the Tower of Babel

Full text

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

Uniform response theory of non-Hermitian systems: Non-Hermitian physics beyond the exceptional point Subhajyoti Bid, Henning Schomerus Abstract Non-Hermitian systems display remarkable response effects that directly reflect a variety of distinct spectral scenarios, such as exceptional points where the eigensystem becomes defective. However, present frameworks treat the different scenarios as separate cases, following the singular mathematical change between different spectral decompositions from one scenario to another. This not only complicates the coherent description near the spectral singularities where the response qualitatively changes, but also impedes the application to practical systems, as the determination of these decompositions is manifestly ill-conditioned. Here we develop a general response theory of non-Hermitian systems that uniformly applies across all spectral scenarios. We unravel this response by formulating a uniform expansion of the spectral quantization condition, as well as a uniform expansion of the Green’s function, where both expansions exclusively involve directly calculable data from the Hamiltonian. This data smoothly varies with external parameters and energy as spectral singularities are approached and attained, and nevertheless captures the qualitative differences of the response in these scenarios. We furthermore present two direct applications of this framework. Firstly, in the context of the quantization condition, we determine the precise conditions for spectral degeneracies of geometric multiplicity greater than unity, as well as the perturbative behavior around these cases. Secondly, in the context of the Green’s function, we formulate a hierarchy of spectral response strengths that varies continuously across all parameter space, and thereby also reliably determines the response strength of exceptional points. Finally, we join both themes, and demonstrate both generally and in concrete examples that the previously inaccessible scenarios of higher geometric multiplicity result in unique variants of super-Lorentzian response. Our approach widens the scope of non-Hermitian response theory to capture all spectral scenarios on an equal and uniform footing, identifies the exact mechanisms that lead to the qualitative changes of physical signatures, and renders non-Hermitian response theory fully applicable to numerical descriptions of practical systems. I Introduction Effectively non-Hermitian Hamiltonians appear naturally in the study of a wide range of open quantum and classical systems Moiseyev 2011 ; Kato 1966 ; Ashida et al. 2020 , including mean-field descriptions of photonic systems with gain and loss Makris et al. 2008 ; Rüter et al. 2010 ; El-Ganainy et al. 2018 ; Cao and Wiersig 2015 , scattering systems Beenakker 1997 ; Guhr et al. 1998 ; Schomerus 2013 , and post-selected measurement protocols Dalibard et al. 1992 ; Carmichael 2009 ; Gopalakrishnan and Gullans 2021 . In the past decade, these systems have gained further attention in the context of non-Hermitian topology Poli et al. 2015 ; Kawabata et al. 2019 ; Bergholtz et al. 2021 ; Okuma and Sato 2023 ; Ota et al. 2020 , in which qualitatively different universality classes can be obtained by the use of symmetry. Non-Hermitian terms in the Hamiltonian greatly enhance the variety of distinct symmetry classes as they make the energy spectrum complex, where the imaginary parts of the energies determine the intensity growth and decay rates of the eigenstates in consideration. Special attention is then drawn to phase transitions in which the spectrum reconfigures due to complex-eigenvalue degeneracies. These degeneracies do not simply replace the band-closing transitions in Hermitian topology. This is because the eigenvectors of non-Hermitian Hamiltonians are not constrained to be orthogonal to each other, so that the generic eigenvalue degeneracies become exceptional points (EPs) where the associated eigenvectors coalesce as well Kato 1966 ; Heiss 2012 ; Heiss 2004 ; Berry 2004 ; Miri and Alu 2019 . Mathematically, these EPs correspond to the problem of degenerate eigenvalues with algebraic multiplicity (the number of eigenvalues that merge, which determines the order of the EP) that differs from the geometric multiplicity (the number of linearly independent eigenvectors, which reduces to a single one for generic EPs). In contrast, Hermitian systems only allow for diabolic points (DPs), where both notions of multiplicity coincide. The EPs themselves exhibit a rich topology Bergholtz et al. 2021 ; Yang et al. 2021 ; Le et al. 2022 ; Ryu et al. 2024 , which changes the nature of phase transitions and further enriches the topological landscape of non-Hermitian systems Malzard et al. 2015 ; Lee 2016 ; Lang et al. 2018 ; Lieu 2018 ; Xue et al. 2020 ; Mostafavi et al. 2020 ; Ghorashi et al. 2021 ; Denner et al. 2021 ; Mandal and Bergholtz 2021 ; Ding et al. 2022 ; Bid et al. 2023 ; Sayyad et al. 2023 ; Dash et al. 2024 ; König et al. 2024 . A central issue in recent research is the manner in which these mathematical characteristics manifest within the system’s physical behavior. These investigations started with the observation that already in the non-degenerate case, the mode-nonorthogonality enhances the sensitivity to static and dynamic perturbations Petermann 1979 ; Siegman 1989 ; Patra et al. 2000 ; Schomerus 2020 , leading to an increased response when compared to standard Breit-Wigner resonance theory Frahm et al. 2000 . The enhancement factor, known as the Petermann factor, diverges at EPs Heiss 2000 ; Berry 2004 . This behavior reflects that mathematically the eigensystem no longer forms a complete set Kato 1966 , which leaves the Hamiltonian non-diagonalizable and hence not amenable to conventional response theory. Resorting instead to the Jordan decomposition (the generalized spectral decomposition based on the Jordan normal form Gantmakher 2000 ), it can be established that the response of the system then changes qualitatively Yoo et al. 2011 ; Heiss 2015 ; Takata et al. 2021 ; Hashemi et al. 2022 ; Schomerus 2022 ; Wiersig 2023a . As a function of energy, this results in super-Lorentzian lineshapes, which become imprinted, e.g., onto the spectrum of quantum-limited noise, while the parametric dependence leads to sensors with power-law transfer functions Wiersig 2014 ; Chen et al. 2017 ; Wiersig 2020 . Overall, one, therefore, encounters different spectral scenarios that are tied to different mathematical descriptions, and give rise to distinct physical behaviors that are intensely studied to the present date (see e.g. Refs. Yang et al. 2023 ; Lai et al. 2019 ; Hodaei et al. 2017 ; Wu et al. 2021 ; Simonson et al. 2022 ). What is missing is a unifying approach that allows to study a given system uniformly across all spectral scenarios. This arises from the reliance on the Jordan decomposition, which has several drawbacks. (i) The Jordan decomposition changes singularly when one steers the system from one spectral scenario to another. Thereby, each scenario is treated as a separate situation, and key characteristics of the system become expressed in terms of the quantities related to the different mathematical descriptions. (ii) The present descriptions are incomplete, as the Jordan decomposition also singularly depends on the geometric multiplicity of the degeneracies, which can exceed one when additional parameters are controlled or suitable symmetries are imposed. These more complicated scenarios quickly proliferate in number, and so do the possible transitions between these scenarios. (iii) The Jordan decomposition is fundamentally ill conditioned, and hence is impractical for numerical applications. In this work, we overcome these conceptual and practical drawbacks by establishing general and exact expansions of the spectral

Read the complete text