pith. sign in

arxiv: 2505.07798 · v5 · submitted 2025-05-12 · 🪐 quant-ph

PT symmetry and the square well potential: Antilinear symmetry rather than Hermiticity in scattering processes

Philip D. Mannheim This is my paper

Pith reviewed 2026-05-22 15:22 UTC · model grok-4.3

classification 🪐 quant-ph
keywords PT symmetrysquare well potentialscattering resonancescomplex energiesantilinear symmetryCPT symmetryHermiticityexceptional points
0
0 comments X

The pith

A real square-well potential has PT and C symmetry, acting Hermitian below the scattering threshold and non-Hermitian above it with complex conjugate energy pairs for resonances.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper shows that the Schrödinger equation for a real square-well potential possesses both C and PT symmetry in the bound and scattering sectors. Below the scattering threshold energies are real, but above it resonances appear as complex conjugate pairs because the scattering amplitude is CPT symmetric. Their interplay enforces probability conservation, so each pair describes one observable resonance rather than two. The Hamiltonian is therefore Hermitian below threshold and non-Hermitian above it. The square well supplies a concrete case where antilinear symmetry is more general than Hermiticity, and at certain potential strengths the threshold branch point becomes an exceptional point.

Core claim

The non-relativistic real potential square-well Schrödinger equation possesses C and PT symmetry in both the bound and scattering sectors, with there being complex conjugate pairs of energy eigenvalue solutions in the scattering sector. The Hamiltonian thus acts as a Hermitian operator below the scattering threshold and as a non-Hermitian one above it. For those values of the potential for which bound states lie right at the top of the well the scattering amplitude threshold branch point is an exceptional point, a characteristic of systems with antilinear symmetry at which there are more independent solutions to the Schrödinger equation than there are eigenstates of a then non-Hermitian. The

What carries the argument

CPT symmetry of the scattering amplitude, which enforces complex conjugate energy pairs to describe resonance excitation and decay while their interplay conserves probability.

If this is right

  • Bound states below threshold have real energies while scattering resonances above threshold have complex conjugate energy pairs.
  • Probability conservation in resonance scattering is maintained by the interplay within each complex conjugate energy pair.
  • Each observable resonance corresponds to one complex energy pair rather than two separate resonances.
  • At specific potential strengths where bound states sit at the top of the well, the scattering threshold branch point becomes an exceptional point.
  • Antilinearity is more general than Hermiticity for real potentials in scattering processes.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same antilinear symmetry argument could apply to other real potentials, changing how resonances are interpreted in scattering theory.
  • Scattering experiments at potentials tuned near threshold might reveal exceptional points through changes in the number of independent solutions.
  • Resonance analysis could shift toward treating each conjugate pair as a single unified excitation-decay process.

Load-bearing premise

The scattering amplitude is CPT symmetric, which then requires complex-conjugate energy pairs whose interplay enforces probability conservation and that each such pair corresponds to only one observable resonance.

What would settle it

An explicit calculation of the scattering amplitude for the square well that finds it is not CPT symmetric or that probability conservation holds with a single complex energy pole rather than a conjugate pair.

read the original abstract

A real potential Hamiltonian has real energy bound states below the scattering threshold and complex energy resonances above it. Scattering states are not square integrable, being instead delta function normalized. This lack of square integrability breaks the connection between Hermiticity and real eigenvalues, to thus allow for real bound state sector eigenvalues and complex scattering sector eigenvalues. When written as contour integrals delta functions take support in the complex plane, with the scattering amplitude being able to take support in the complex plane too. However, the scattering amplitude is CPT symmetric. For resonance scattering this antilinear symmetry requires the presence of a complex conjugate pair of energies, one to describe the excitation of the resonance and the other to describe its decay, with it being their interplay that enforces probability conservation. Each complex pair of energy eigenvalues corresponds to only one observable resonance not two, to thus modify the standard pure decaying complex energy pole discussion of resonances. We show that the non-relativistic real potential square-well Schr\"odinger equation possesses C and PT symmetry in both the bound and scattering sectors, with there being complex conjugate pairs of energy eigenvalue solutions in the scattering sector. The Hamiltonian thus acts as a Hermitian operator below the scattering threshold and as a non-Hermitian one above it. For those values of the potential for which bound states lie right at the top of the well the scattering amplitude threshold branch point is an exceptional point, a characteristic of systems with antilinear symmetry at which there are more independent solutions to the Schr\"odinger equation than there are eigenstates of a then non-Hermitian Hamiltonian. The square well provides an explicit realization of how antilinearity is more general than Hermiticity.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The manuscript claims that the real square-well Schrödinger Hamiltonian is Hermitian for bound states below the scattering threshold but non-Hermitian above it. It argues that delta-function normalization of scattering states breaks the usual link between Hermiticity and real eigenvalues, allowing complex-conjugate energy pairs in the scattering sector; CPT symmetry of the scattering amplitude then requires these pairs to enforce probability conservation, with each pair corresponding to a single observable resonance. The square well is said to realize C and PT symmetry in both sectors, with antilinearity being more general than Hermiticity; for potentials where a bound state sits at threshold, the branch point becomes an exceptional point.

Significance. If the central claim were correct it would offer a new conceptual framing of resonances via antilinear symmetries that unifies bound and scattering sectors. The manuscript does not, however, supply the operator-theoretic or contour-integral derivations needed to substantiate the switch from Hermitian to non-Hermitian behavior, nor does it contain numerical checks of the claimed symmetry properties or exceptional-point structure.

major comments (3)
  1. [Abstract and §1] Abstract and §1: The assertion that the Hamiltonian 'acts as a Hermitian operator below the scattering threshold and as a non-Hermitian one above it' is not supported by the spectral theory of the Schrödinger operator. For real V(x) the operator −d²/dx² + V(x) is self-adjoint on L²(ℝ) (or on the appropriate domain for the square well) with absolutely continuous spectrum [0,∞); complex resonances appear only as poles of the meromorphic continuation of the resolvent or S-matrix, not as eigenvalues of the Hamiltonian itself.
  2. [Abstract] Abstract: The claim that 'delta-function normalization breaks the connection between Hermiticity and real eigenvalues' conflates the continuous spectrum (which remains real) with the analytic continuation that produces resonances. Standard texts (e.g., Newton, Scattering Theory of Waves and Particles) show that probability conservation follows from unitarity of the S-matrix on the real axis without requiring the Hamiltonian to become non-Hermitian.
  3. [Abstract] Abstract and discussion of exceptional points: The statement that the threshold branch point becomes an exceptional point 'at which there are more independent solutions to the Schrödinger equation than there are eigenstates of a then non-Hermitian Hamiltonian' is not accompanied by an explicit demonstration that the operator ceases to be self-adjoint or that the algebraic multiplicity exceeds the geometric multiplicity at that parameter value.
minor comments (2)
  1. The manuscript would benefit from explicit reference to the standard treatment of resonances as poles of the S-matrix continuation rather than as eigenvalues.
  2. Notation for the contour integrals that place the delta functions in the complex plane should be defined more precisely, including the choice of contour and its relation to the physical sheet.

Simulated Author's Rebuttal

3 responses · 1 unresolved

We thank the referee for the careful and detailed report. We address each major comment below, defending our interpretation of antilinear symmetry in the scattering sector while acknowledging standard spectral theory where appropriate.

read point-by-point responses

  1. Referee: [Abstract and §1] The assertion that the Hamiltonian 'acts as a Hermitian operator below the scattering threshold and as a non-Hermitian one above it' is not supported by the spectral theory of the Schrödinger operator. For real V(x) the operator −d²/dx² + V(x) is self-adjoint on L²(ℝ) (or on the appropriate domain for the square well) with absolutely continuous spectrum [0,∞); complex resonances appear only as poles of the meromorphic continuation of the resolvent or S-matrix, not as eigenvalues of the Hamiltonian itself.

    Authors: We agree that the differential operator with real potential is self-adjoint on the square-integrable domain with real continuous spectrum. Our claim concerns the effective description in the scattering sector, where states are delta-normalized rather than square-integrable. This normalization, combined with analytic continuation to complex resonance poles, permits complex eigenvalues whose conjugate pairs are enforced by CPT symmetry of the scattering amplitude. The framework does not alter the self-adjointness on the real axis but shows how antilinearity becomes the operative symmetry above threshold, unifying bound and scattering sectors. revision: no

  2. Referee: [Abstract] The claim that 'delta-function normalization breaks the connection between Hermiticity and real eigenvalues' conflates the continuous spectrum (which remains real) with the analytic continuation that produces resonances. Standard texts (e.g., Newton, Scattering Theory of Waves and Particles) show that probability conservation follows from unitarity of the S-matrix on the real axis without requiring the Hamiltonian to become non-Hermitian.

    Authors: We do not contest the reality of the continuous spectrum or the derivation of S-matrix unitarity on the real axis. Our focus is on the resonances as complex poles, where delta-function normalization allows treatment of these poles as eigenvalues in the scattering sector. CPT symmetry then requires conjugate pairs whose interplay enforces probability conservation for each observable resonance. This provides a symmetry-based account complementary to the standard unitarity argument. revision: no

  3. Referee: [Abstract] The statement that the threshold branch point becomes an exceptional point 'at which there are more independent solutions to the Schrödinger equation than there are eigenstates of a then non-Hermitian Hamiltonian' is not accompanied by an explicit demonstration that the operator ceases to be self-adjoint or that the algebraic multiplicity exceeds the geometric multiplicity at that parameter value.

    Authors: For the exactly solvable square well, the threshold case produces a branch point in the scattering amplitude at which independent solutions coalesce, consistent with exceptional-point behavior under antilinear symmetry. While a complete algebraic-multiplicity analysis of the non-self-adjoint operator is not supplied, the explicit wave-function matching and S-matrix poles demonstrate the degeneracy and the transition to non-Hermitian-like features. We are prepared to add a short clarifying calculation in revision. revision: partial

standing simulated objections not resolved
  • The manuscript does not supply the full operator-theoretic or contour-integral derivations for the Hermitian-to-non-Hermitian transition or numerical checks of the symmetry properties.

Circularity Check

0 steps flagged

No significant circularity; derivation rests on external standard normalizations and symmetries.

full rationale

The paper's chain begins from the standard delta-function normalization of non-square-integrable scattering states and the CPT symmetry of the S-matrix, both of which are external inputs from conventional quantum mechanics rather than derived or fitted within the work. The claim that this normalization 'breaks the connection between Hermiticity and real eigenvalues' is presented as a direct logical consequence, not a self-referential redefinition or a prediction that reduces to the input by construction. No self-citations are load-bearing for the central premise, no parameters are fitted to a subset and then relabeled as predictions, and the square-well example is used illustratively without smuggling an ansatz or renaming a known result as a new unification. The overall argument remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on standard quantum-mechanical normalization and the CPT symmetry of the scattering amplitude; no free parameters, new entities, or ad-hoc axioms are introduced in the abstract.

axioms (2)
  • standard math Scattering states are delta-function normalized and not square-integrable.
    Standard property of continuum states in quantum mechanics invoked to break the Hermiticity-real-eigenvalue link.
  • domain assumption The scattering amplitude is CPT symmetric.
    Invoked to require complex-conjugate energy pairs for resonances.

pith-pipeline@v0.9.0 · 5833 in / 1315 out tokens · 38534 ms · 2026-05-22T15:22:20.993588+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

38 extracted references · 38 canonical work pages · 1 internal anchor

  1. [1]

    =E 0, ℏ2K1K2 m = Γ, ℏ2 2m(k2 1 −k 2

    work page

  2. [2]

    There is one key difference between the bound state and scattering state solutions that we have found

    =E 0 −V 0, ℏ2k1k2 m = Γ.(5.11) Satisfying the boundary conditions atr=aleads to tan((K1 ±iK 2)a) = K1 ±iK 2 i(k1 ±ik 2) ,sin((K 1 ±iK 2)a) = K1 ±iK 2 [(K1 ±iK 2)2 −(k 1 ±ik 2)2]1/2 = E0 ±iΓ V0 1/2 ,(5.12) which we recognize as (5.6) as evaluated above threshold. There is one key difference between the bound state and scattering state solutions that we hav...

    work page

  3. [3]

    P. D. Mannheim, J. Phys. A: Math. Theor.51, 315302 (2018)

    work page 2018

  4. [4]

    P. D. Mannheim, Phys. Rev. D112, L031903 (2025)

    work page 2025

  5. [5]

    Sinclair, D

    J. Sinclair, D. Angulo, K. Thompson, K. Bonsma-Fisher, A. Brodutch and A. M. Steinberg, PRX Quantum3, 010314 (2022)

    work page 2022

  6. [6]

    Angulo, K

    D. Angulo, K. Thompson, V-M. Nixon, A. Jiao, H. M. Wiseman and A. M. Steinberg, arXiv:2409.03680 [quant-ph)

    work page arXiv

  7. [7]

    E. P. Wigner, J. Math. Phys.1, 409 (1960)

    work page 1960

  8. [8]

    M. A. Sim´ on, A. Buendia, A. Kiely, A. Mostafazadeh and J. G. Muga, Phys. Rev. A99, 052110 (2019)

    work page 2019

  9. [9]

    Some of the analysis that we present here duplicates that of [1] and the short letter in [2] and is included here as it provides the background needed for this paper, and enables this paper to be self-contained and complete

    work page

  10. [10]

    E. P. Wigner, Phys. Rev. D98, 145 (1955)

    work page 1955

  11. [11]

    Messiah,Quantum MechanicsVol

    A. Messiah,Quantum MechanicsVol. 1 (North Holland Publishing Company, Amsterdam 1964)

    work page 1964

  12. [12]

    P. D. Mannheim, arXiv:2602.17589 [quant-ph]

    work page arXiv

  13. [13]

    Inserting the real parts ofKandkinto (5.2) gives tanδ= 1.748, which is far from the infinite tan(π/2)

    Settingγ= 1,a= 1 andV 0 = 16, from (5.7) we obtainµ= 7.661,ν= 1.288,E 0 = 57.025, Γ = 19.738,K= 7.661±1.288i, k= 6.579±1.500i. Inserting the real parts ofKandkinto (5.2) gives tanδ= 1.748, which is far from the infinite tan(π/2). From (5.2) the realKandkthat would give tanδ=∞, i.e., giveβtanα=−1, areK= 6.045,k= 4.532, withE 0 = 36.542. Withβ=−1/tanα, atk=...

    work page

  14. [14]

    C. M. Bender and S. Boettcher, Phys. Rev. Lett.80, 5243 (1998)

    work page 1998

  15. [15]

    C. M. Bender, S. Boettcher and P. N. Meisinger, J. Math. Phys.40, 2201 (1999)

    work page 1999

  16. [16]

    Mostafazadeh, J

    A. Mostafazadeh, J. Math. Phys.43, 205 (2002); J. Math. Phys.43, 2814 (2002); J. Math. Phys.43, 3944 (2202)

    work page 2002

  17. [17]

    C. M. Bender, D. C. Brody and H. F. Jones, Phys. Rev. Lett.89, 270401 (2002)

    work page 2002

  18. [18]

    C. M. Bender, Rep. Prog. Phys.70, 947 (2007). 22

    work page 2007

  19. [19]

    K. G. Makris, R. El-Ganainy, D. N. Christodoulides and Z. H. Musslimani, Phys. Rev. Lett.100, 103904 (2008)

    work page 2008

  20. [20]

    C. M. Bender and P. D. Mannheim, Phys. Rev. Lett.100, 110402 (2008)

    work page 2008

  21. [21]

    C. M. Bender and P. D. Mannheim, Phys. Rev. D78, 025022 (2008)

    work page 2008

  22. [22]

    A. Guo, G. J. Salamo, D. Duchesne, R. Morandotti, M. Volatier-Ravat, V. Aimez, G. A. Siviloglou and D. N. Christodoulides, Phys. Rev. Lett.103, 093902 (2009)

    work page 2009

  23. [23]

    C. M. Bender and P. D. Mannheim, Phys. Lett. A374, 2616 (2010). [22]Special issue on quantum physics with non-Hermitian operators, C. Bender, A. Fring, U. G¨ unther, and H. Jones (Guest Editors) J. Phys. A: Math. Theor.45, 444001 - 444036 (2012). [23]Theme issue on PT quantum mechanics, C. M. Bender, M. DeKieviet, and S. P. Klevansky (Guest Editors) Phil....

    work page 2010

  24. [24]

    El-Ganainy, K

    R. El-Ganainy, K. G. Makris, M. Khajavikhan, Z. H. Musslimasni, S. Rotter and D. M. Christodoulides, Nature Phys.14, 1119 (2018)

    work page 2018

  25. [25]

    C. M. Bender,PT Symmetry in Quantum And Classical Physics, World Scientific Press, Singapore (2018)

    work page 2018

  26. [26]

    Fring, F

    A. Fring, F. Correa and C. M. Bender, Proceedings of Virtual seminar series on Pseudo-Hermitian Hamiltonians in Quantum Physics, Journal of Physics Conference Series, Vol.2038(2021)

    work page 2038

  27. [27]

    P. A. M. Dirac, Proc. Roy. Soc. A180, 1 (1942)

    work page 1942

  28. [28]

    Pauli, Rev

    W. Pauli, Rev. Mod. Phys.15, 175 (1943)

    work page 1943

  29. [29]

    P. D. Mannheim, Int. J. Mod. Phys. D33, 2441016 (2024)

    work page 2024

  30. [30]

    P. D. Mannheim, Phil. Trans. R. Soc. A371, 20120060 (2013)

    work page 2013

  31. [31]

    T. D. Lee and G. C. Wick, Nucl. Phys. B9, 209 (1969); Nucl. Phys. B10, 1 (1969); Phys. Rev. D2, 1033 (1970)

    work page 1969

  32. [32]

    In their original approach Dirac and Pauli introduced a unitarity-violating negative Hilbert space metric and a propagator of the form 1/(k 2 −M 2 1 )−1/(k 2 −M 2 2 ) with a relative minus sign, as it would be more convergent at largek 2 than the standard single pole 1/(k 2 −M 2 1 ). To avoid the indefinite metric Lee and Wick introduced the complex conju...

    work page

  33. [33]

    Given the presence of the causalθ(t) factor in (6.9) an excited state does not emit a photon before the ground state absorbs it, i.e., the time advance is not bigger than the time delay

    work page

  34. [34]

    In that effect the center of a wave packet approaching a well (or barrier) travels on the classical trajectory

    TheP T-symmetry time advance discussed here is not the same as a familiar time advance effect that occurs in quantum mechanics. In that effect the center of a wave packet approaching a well (or barrier) travels on the classical trajectory. However, the wings of the wave packet not only spread out they can start to deform due the presence of the well befor...

    work page

  35. [35]

    ThusM(s >1) is a Hermitian matrix as written in a skew basis rather than an orthogonal one

    For the real eigenvalue case withs >1 we set sinα= (s 2 −1) 1/2/s,A= [(1 + sinα)/2 sinα] 1/2,B= [(1−sinα)/2 sinα] 1/2, S=Aσ 0 +Bσ 2, and find [1] that under a similarity transformationM(s >1) transforms into the HermitianSM(s > 1)S−1 =σ 0 +σ 1 tanα. ThusM(s >1) is a Hermitian matrix as written in a skew basis rather than an orthogonal one

    work page

  36. [36]

    For the complex conjugate eigenvalue case withs <1 we can diagonalizeM(s <1) viaSM(s <1)S −1 whereS= coshα+σ 2 sinhαwith tanh(2α) =s, and obtain (6.4) as written withE 0 = 1 and Γ = (1−s 2)1/2

    work page

  37. [37]

    P. D. Mannheim and A. Davidson, arXiv: hep-th/0001115; Phys. Rev. A71, 042110 (2005)

    work page internal anchor Pith review Pith/arXiv arXiv 2005

  38. [38]

    It was through study of systems such as these that a role forP Tsymmetry in quantum theory was identified, with any theory with an antilinear symmetry then generically being referred to as aP Ttheory

    work page