Geometric and Topological Obstructions to Hermitianization in Quasi-Hermitian Quantum Systems

This paper establishes that while quasi-Hermitian quantum systems can be locally mapped to Hermitian ones, their global dynamical equivalence is obstructed by geometric curvature and topological holonomies in parameter space, which determine whether intrinsic non-Hermitian features persist.

The Gist

The Big Picture: Turning a "Tricky" System into a "Normal" One

Imagine you are trying to navigate a complex, strange city (a Quasi-Hermitian Quantum System). In this city, the rules of the road are weird. Distances aren't measured with a standard ruler; instead, you have a special, flexible measuring tape that stretches and shrinks depending on where you are. This makes calculating things like energy and movement very difficult.

Physicists have a trick to make this city easier to understand: they want to map it onto a standard, normal city (a Hermitian System) where the rules are simple, distances are fixed, and everything behaves predictably.

To do this, they use a "translation tool" called a Similarity Transformation. Think of this tool as a pair of special glasses or a map converter. If you put the glasses on, the weird city looks exactly like the normal city.

The Problem:
The paper asks a crucial question: Can we always wear these glasses and see the normal city clearly, no matter where we walk?

The authors discovered that sometimes, you cannot put on the glasses and see the whole city at once. There are two specific "roadblocks" that prevent this translation from working globally. They call these Geometric Obstructions and Topological Obstructions.

---

Obstruction #1: The Geometric Hump (Curvature)

The Analogy:
Imagine you are walking on the surface of a sphere (like a beach ball). You try to draw a grid of straight lines (latitude and longitude) to map the surface.

• If you walk in a small circle, you can draw a perfect grid.
• But if you try to draw a grid that covers the entire sphere without it getting messy or overlapping, you fail. The surface is "curved." If you try to flatten a globe onto a piece of paper, the map gets distorted.

What the Paper Says:
In the quantum system, the "special measuring tape" (called the metric) creates a kind of curvature in the mathematical space.

• The Result: If this curvature is not zero, you cannot create a single, consistent map (a global transformation) that turns the whole weird system into a normal one.
• The Symptom: If you walk in a circle in the original "weird" system and come back to the start, everything looks the same. But if you try to translate that path into the "normal" system, the path might not close! You might end up at a slightly different spot than where you started. The "normal" system becomes non-periodic (it doesn't repeat neatly) even though the original one did.

In short: The terrain is too bumpy to flatten out completely.

---

Obstruction #2: The Topological Hole (The Donut Effect)

The Analogy:
Now, imagine the surface is perfectly flat (no hills or bumps), but it has a hole in the middle, like a donut or a life preserver.

• You can walk around the donut.
• If you walk around the hole, you can't shrink your path down to a single point without crossing the hole.
• Imagine you are carrying a compass. As you walk around the hole, the compass needle might slowly rotate. When you return to your starting point, the compass is pointing in a different direction than when you left, even though the ground was perfectly flat.

What the Paper Says:
Even if the "curvature" is zero (the ground is flat), the shape of the space can still cause problems.

• The Result: If the space has a "hole" (a non-contractible loop), the translation tool (the glasses) might twist as you walk around it.
• The Symptom: When you return to the start, the translation tool might be "flipped" or rotated. It's like if you walked around a pole and your glasses turned upside down. Because of this twist, you cannot define a single, consistent map for the whole system. The "normal" system you see through the glasses will have a different "twist" or phase than the original system.

In short: The space has a hole, and walking around it twists your translation tool, making a global map impossible.

---

The Three Examples the Authors Used

To prove these ideas, the authors built three specific models:

1. The Easy Case (No Obstruction):

   • Scenario: A system where the "measuring tape" is simple and the space has no holes.
   • Outcome: You can wear the glasses perfectly. The weird system maps 100% to a normal system. Everything works smoothly.

2. The Curved Case (Geometric Obstruction):

   • Scenario: A system on a disk (a flat circle) where the "measuring tape" creates a hump (curvature) in the middle.
   • Outcome: You can only map the system perfectly if you walk along a very specific, special circle where the math lines up perfectly. If you walk on any other circle, the map breaks. The "normal" system becomes a twisted, non-repeating mess.

3. The Holey Case (Topological Obstruction):

   • Scenario: A system on a ring (an annulus) with a hole in the middle. The ground is perfectly flat (no curvature).
   • Outcome: Even though the ground is flat, walking around the hole twists the translation tool. The "normal" system you see has a different phase (a different "twist") than the original. You cannot make a single map that works for the whole ring.

The Bottom Line

The paper establishes that you cannot always assume a "weird" quantum system is just a "normal" system in disguise.

• Sometimes, the shape of the space (curvature) prevents the translation.
• Sometimes, the holes in the space (topology) prevent the translation.

If either of these obstructions exists, the system has intrinsic non-Hermitian features. It is fundamentally different from a standard quantum system, and trying to force it to look like a normal one will result in a broken or twisted map.

Technical Summary

Technical Summary: Geometric and Topological Obstructions to Hermitianization in Quasi-Hermitian Quantum Systems

Problem Statement
Quasi-Hermitian quantum systems, including those with parity-time ($PT$) symmetry, possess entirely real energy spectra and admit a consistent probabilistic interpretation through a positive-definite metric operator $\eta$. A standard strategy to analyze these systems is to map them to equivalent Hermitian systems via a similarity transformation $S$ such that $\eta = S^\dagger S$ and $\tilde{H} = SHS^{-1}$ is Hermitian. While an instantaneous, algebraic Hermitianization is always possible locally for a fixed parameter set, the paper addresses a critical gap: the existence of a global, single-valued similarity transformation that preserves dynamical equivalence. Specifically, for the mapped system to obey the standard Schrödinger equation, the transformation $S$ must satisfy a "proper" condition ($\dot{S}^\dagger S = S^\dagger \dot{S}$). The authors investigate whether such a global, proper $S$ always exists and identify the conditions under which it fails, leading to intrinsic non-Hermitian features that cannot be removed by a global transformation.

Methodology
The authors develop a framework based on the geometry of the metric-induced gauge connection.

1. Proper Similarity Transformation: They define a proper transformation $S_p$ that satisfies the dynamical compatibility condition. They show that locally, $S_p$ can be constructed as $S_p = U\sqrt{\eta}$, where $U$ is a unitary operator determined by a first-order differential equation involving the metric.
2. Integrability and Obstructions: The global existence of a single-valued $S_p$ is linked to the integrability of the differential equation governing $U$. The authors introduce a gauge connection $G_\mu = \frac{1}{2}[\partial_\mu \sqrt{\eta}, \sqrt{\eta}^{-1}]$ derived from the metric.
3. Curvature and Holonomy: They analyze two distinct types of obstructions:
   • Geometric Obstruction: Arising from the non-zero curvature ($F^G_{\mu\nu} \neq 0$) of the connection $G_\mu$. This prevents the existence of a consistent local frame globally.
   • Topological Obstruction: Arising from non-trivial holonomies (Wilson loops) around non-contractible loops in the parameter space, even when the curvature vanishes ($F^G_{\mu\nu} = 0$).
4. Berry Phase Analysis: The authors derive the relationship between the Berry connection of the quasi-Hermitian system and the mapped Hermitian system. They show that the difference in Berry curvatures is directly proportional to the curvature of the gauge connection, providing a diagnostic for the obstruction.
5. Explicit Examples: Three representative models are constructed to illustrate these concepts: a system with no obstructions, a system on a disk with non-zero curvature, and a system on an annulus with vanishing curvature but non-trivial topology.

Key Contributions and Results

• Identification of Two Obstructions: The paper rigorously identifies and distinguishes between geometric obstructions (due to non-zero gauge curvature) and topological obstructions (due to non-trivial Wilson loops).
• Explicit Criteria:
  • A geometric obstruction exists if the curvature $F^G_{\mu\nu} = \partial_\mu G_\nu - \partial_\nu G_\mu - [G_\mu, G_\nu]$ is non-zero. In this case, no global single-valued $S_p$ exists.
  • A topological obstruction exists if the Wilson loop $W(C) = \mathcal{P} \exp(-\oint_C G_\mu dR^\mu)$ around a non-contractible loop is not the identity, even if the curvature vanishes everywhere.
• Physical Manifestations:
  • Non-Periodicity: In the presence of a geometric obstruction, a closed loop in the parameter space of the original quasi-Hermitian system maps to an open path in the equivalent Hermitian system. The mapped Hamiltonian $\tilde{H}$ fails to be periodic unless specific quantization conditions are met.
  • Berry Phase Mismatch: The paper demonstrates that the geometric and topological obstructions lead to discrepancies between the Berry phases of the quasi-Hermitian system and the locally mapped Hermitian system. In the topological case (Example 3), the quasi-Hermitian system yields a zero Berry phase for a loop enclosing a hole, while the mapped Hermitian system yields a non-zero phase ($-\pi$) due to the multi-valued nature of the transformation.
• Equivalence of Conditions: The authors prove that the integrability condition for the transformation $S_p$ (expressed via an operator $K_\mu$) is equivalent to the vanishing of the curvature $F^G_{\mu\nu}$, establishing a direct link between the non-integrability of the connection and the difference in Berry curvatures.

Significance
The paper establishes a geometric and topological foundation for the Hermitianization of quasi-Hermitian systems. It clarifies that while local equivalence to a Hermitian system is guaranteed, global equivalence is not. The results provide precise criteria for determining when a quasi-Hermitian system can be globally reduced to a Hermitian one and when intrinsic non-Hermitian features persist. This work is essential for correctly interpreting geometric phases, quantum geometric tensors, and other invariants in non-Hermitian physics, particularly in systems with non-trivial parameter space topologies or non-flat metric-induced connections. The findings explain the origin of "open-path" phenomena observed in time-dependent quasi-Hermitian quantum mechanics, attributing them to the curvature and topology of the metric-induced connection rather than a failure of the Hermitianization strategy itself.