SciPost Submission Page
Symmetryprotected exceptional and nodal points in nonHermitian systems
by Sharareh Sayyad, Marcus Stålhammar, Lukas Rodland, Flore K. Kunst
This is not the latest submitted version.
This Submission thread is now published as
Submission summary
Authors (as registered SciPost users):  Sharareh Sayyad · Marcus Stålhammar 
Submission information  

Preprint Link:  scipost_202304_00017v1 (pdf) 
Date submitted:  20230417 09:35 
Submitted by:  Sayyad, Sharareh 
Submitted to:  SciPost Physics 
Ontological classification  

Academic field:  Physics 
Specialties: 

Approach:  Theoretical 
Abstract
One of the unique features of nonHermitian~(NH) systems is the appearance of NH degeneracies known as exceptional points~(EPs). The extensively studied defective EPs occur when the Hamiltonian becomes nondiagonalizable. Aside from this degeneracy, we show that NH systems may host two further types of nondefective degeneracies, namely, nondefective EPs and ordinary~(Hermitian) nodal points. The nondefective EPs manifest themselves by i) the diagonalizability of the NH Hamiltonian at these points and ii) the nondiagonalizability of the Hamiltonian along certain intersections of these points, resulting in instabilities in the Jordan decomposition when approaching the points from certain directions. We demonstrate that certain discrete symmetries, namely paritytime, parityparticlehole, and pseudoHermitian symmetry, guarantee the occurrence of both defective and nondefective EPs. We extend this list of symmetries by including the NH timereversal symmetry in twoband systems. Twoband and fourband models exemplify our findings. Through an example, we further reveal that ordinary nodal points may coexist with defective EPs in NH models when the above symmetries are relaxed.
Current status:
Submission & Refereeing History
You are currently on this page
Reports on this Submission
Anonymous Report 2 on 2023623 (Invited Report)
 Cite as: Anonymous, Report on arXiv:scipost_202304_00017v1, delivered 20230623, doi: 10.21468/SciPost.Report.7388
Strengths
1 The technical computations are clearly exposed and easy to follow.
2 The studies of degeneracies in nonHermitian systems has clearly become very relevant for experiments. This paper brings a clearer mathematical understanding to the nuances of their properties.
3 A formal characterization of the NDEP would be appreciated.
Weaknesses
1 In my opinion, the paper fails to properly introduce the different types of degeneracies in nonHermitian systems. A short,mathematical definition (with an example for each) in Introduction or Appendix would greatly simplify the reading of the paper. I , in particular, would stress the importance of a more explicit definition of defective vs nondefective exceptional points early in the paper (it is barely half a sentence currently).
2 Similarly, the authors fail to emphasize the relevance of the distinction between defective and nondefective EP, in terms of physics.
Report
The paper "Symmetryprotected exceptional and nodal points in nonHermitian systems" gives generic criterion of existence for defective and nondefective highdimensional exceptional points in general n bands models.
Given the relevance of the nonHermitian descriptions of experiments and the possibilities offered by exceptional points. the manuscript appears to be relevant to a general public.
While the paper is globally well presented, it is hard to follow, especially for non specialists. I also think there are several important typos. Consequently, I would only recommend the publication in SciPost after modifications.
Given the mistakes I found, I also strongly recommend a careful rereading of the manuscript as it is not impossible I missed others.
I had a few questions, in addition to the changes I would like to see listed below.
1 Nondefective exceptional point are characterized as diagonalizable degenerate points in the neighborhood of which defective EP exists.
Is there a form of (topological) invariant/signature one can derive to characterize them instead?
2 These NDEP split two manifolds/lines of DEP. Given this structure, a property should abruptly change when following these lines of DEP through the NDEP. At first sight, I expect that some properties of the defective eigenstates dramatically change (typically, handedness in the example you have) . Is this intuition correct and general? Can you take advantage of that to explain the resilience of the NDEP in high dimensions?
Requested changes
1 Give concrete and explicit mathematical definitions of the different type of degeneracies discussed in the manuscript, preferably in Section II (or in an Appendix for examples) and stressing the differences.
2End of page 4, you discuss $TRS^\dagger$. You claim that "it enforces all symmetric parts of d to be 0". It does not seem to be the case for $d_{x, I}$, $d_{y, R}$ and $d_{z, I}$ which should be symmetric. Following that it seems that the corresponding result in the table is incorrect.
3 The limits in Eq. 15 appear ill defined: the limit on $k_y$ is only well defined if $k_z^2 > k_x^2$ and then $k_z$ is sent to 0 while $k_x$ is kept finite. The limit on $k_z$ does not appear to be necessary to show the desired property: fixing $k_y^2 = k_z^2  k_x^2$ is enough.
4 Fig. 2a: the spectrum is not symmetric under kx > kx. Given the form and the parameter, there seems to be a mistake.
5 Could you clarify the reason of the double< degeneracy in your 4 band model.
6 The spectrum in Fig. 4
6 Could you develop/clarify and illustrate in Section 4 the experimental signature of the presence of nondefective EP
Minor points
7 Fig. 2 and 3 should be made larger. In both cases, c) and d) are barely understandable. Given that both Figs describe traceless models with only 2 effective bands (with the double degeneracy for Fig. 4), I would recommend plotting a pcolor map (or something similar) of one of the bands only for all of these graphs.
If the authors want to stress square root profiles (or other), then showing a cut would probably be enough.
8 Eq. 17 and 18 are valid for $\tilde{k}_y = \sqrt{k_z^2k_x^2}$. It is not very clearly specified. Also, I seem to find the denominator to be $\tilde{k}_x + \tilde{k}_z$, though it might be a question of convention.
9 Eq. 21 and 22: I also have a minus sign in front of kz.
10 Eq.23 and 24: The two limits are again not really necessary.
Anonymous Report 1 on 2023619 (Invited Report)
 Cite as: Anonymous, Report on arXiv:scipost_202304_00017v1, delivered 20230619, doi: 10.21468/SciPost.Report.7377
Strengths
1.
Report
Dear Editor,
hereby I submit my report on "Symmetryprotected exceptional and nodal points in nonHermitian systems" by Sharareh Sayyad, Marcus Stalhammar, Lukas Rodland, Flore K. Kunst.
The authors studies nonHermitian degeneracies in nonHermitian systems subjected to symmetries. In particular, the authors address defective, nondefective and ordinary Hermitian degeneracies. Of great importance is the recipe to distinguish the three types of degeneracies in nonHermitian setups, which is of importance towards the detecting of exceptional points and their potential applications. Furthermore, I would like to highlight that the manuscript is wellwritten and the messages clearly exposed. For these reasons I recommend its acceptance in SciPost after addressing the minor comments listed below, which might be useful to further improve the manuscript:
1. In page 10 below equation 29, in the sentence starting with "This can be seen from Fig. 4(b)..". I have the feeling that it should be Fig.4(c).
2. I would recommend the authors to add additional references, highlighting the immense efforts of several groups. For instance, see following suggestions:
2.1 Together with Refs. [15], the authors could incorporate the first nonHermitian studies on a consented matter matter junction: JETP Lett. 94, 693 (2012); Scientific Reports 6, 21427 (2016).
2.2 Together with Refs. [619]: Phys. Rev. B 99, 165145 (2019); Phys. Rev. Research 4, L012006 (2022)
2.3 Together with Refs.[2838], I recommend adding: Phys. Rev. B 97, 014512 (2018); Phys. Rev. B 107, 104515 (2023); Phys. Rev. B 107, 115146 (2023); Proc. Int. Conf. on Strongly Correlated Electron Systems (SCES2019) (Physical Society of Japan, 2019) Chap. 30, p. 011098.
3. In section 4, the authors discuss about identifying exceptional points in experiments. Even though the authors address some relevant systems, I believe their manuscript could have a bigger impact if they also discuss nonHermitian systems emerging from material junctions. The material junctions are very relevant in transport experiments, see e.g., Nat. Rev. Mater. 6, 944 (2021), and represent a natural platform for nonHermitian physics. Examples of nonHermitian physics in junctions have been initially reported in JETP Lett. 94, 693 (2012); Scientific Reports 6, 21427 (2016). Later, more interesting studies were reported which further support the importance of nonHermitian physics in material junctions, see e.g., Phys. Rev. B 87, 235421 (2013).; Phys. Rev. Research 1, 012003(R) (2019); Phys. Rev. B 103, 235438 (2021); J. Phys.: Condens. Matter 35, 254002 (2023).
Please, let me know if I can be of further assistance.
Author: Sharareh Sayyad on 20230913 [id 3977]
(in reply to Report 1 on 20230619)We would like to thank the referee for the submitted report. We believe that the feedback has helped us improve the manuscript, in particular, when in relation to applications. Attached, we provide answers to the questions raised by the referees with the original question included in teal.
Author: Sharareh Sayyad on 20230913 [id 3978]
(in reply to Report 2 on 20230623)We would like to thank the referee for the submitted report. We believe that the feedback has helped us improve the manuscript, in particular, when in relation to applications. Attached, we provide answers to the questions raised by the referee with the original question included in teal.
Attachment:
reply_referee2.pdf