Non-Hermitian skin effect and electronic nonlocal transport

This paper demonstrates that the non-Hermitian skin effect in a Rashba nanowire coupled to a ferromagnetic lead can be detected through nonreciprocal nonlocal conductance, offering a new transport spectroscopy method to probe non-Hermitian phenomena and explaining the parameter-space shifts of exceptional points under open boundary conditions.

The Gist

The Big Idea: A One-Way Street for Electrons

Imagine you are running a marathon on a track. In a normal world (what physicists call "Hermitian" physics), if you run from the start line to the finish line, you use the same amount of energy as if you ran from the finish line back to the start. The track is fair and symmetrical.

But this paper describes a magical, broken track (a "Non-Hermitian" system). On this track, the rules of physics are slightly different because the track is "leaking" energy to the outside world.

The authors discovered that on this broken track, electrons don't just run; they pile up at one specific end of the track. If you try to push them from the other end, they get stuck or lost. This phenomenon is called the Non-Hermitian Skin Effect (NHSE).

The Setup: The "Leaky" Nanowire

The scientists built a theoretical model of a tiny wire (a nanowire) made of special materials. Here is how they made it "leaky":

1. The Wire: It's a semiconductor wire with a special property called "Rashba spin-orbit coupling." Think of this as a rule that says: "If you run to the right, you must spin clockwise. If you run to the left, you must spin counter-clockwise."
2. The Magnet: They attached a ferromagnetic lead (a magnet) to the wire. This magnet acts like a sponge that soaks up electrons, but only if they are spinning in a specific direction.
3. The Trap: Because of the wire's rules, electrons running one way spin in a way that the sponge loves (they get soaked up/absorbed). Electrons running the other way spin in a way the sponge ignores (they pass through safely).

The Discovery: The "Skin" Effect

When they ran a simulation, they found something weird happening:

• The "Skin" Piles Up: The electrons that can be absorbed by the sponge don't just disappear; they get pushed and crowded against the end of the wire where the sponge is. They form a "skin" of electrons stuck at the boundary.
• The One-Way Street:
  • Scenario A (Left to Right): If you push electrons from the "safe" side, they flow through the wire easily.
  • Scenario B (Right to Left): If you push electrons from the "sponge" side, they get stuck at the edge and can't get through.

This creates a non-reciprocal effect. It's like a door that opens easily if you push from the outside, but is jammed shut if you try to push from the inside.

How They Proved It: The "Symmetric vs. Asymmetric" Test

The authors proposed a way to see this in a real lab experiment using two types of measurements:

1. Local Conductance (The "Mirror" Test):

   • They measured how well electricity flows out of the wire at the exact spot where they put it in.
   • Result: It looked the same on both ends. It was symmetrical.
   • Analogy: If you shout into a room with a leaky wall, the echo you hear right next to your mouth sounds the same whether you are on the left or right side of the room. The "local" view hides the problem.

2. Non-Local Conductance (The "Cross-Talk" Test):

   • They put the voltage source on the Left and measured the current coming out the Right (and vice versa).
   • Result: This is where the magic happened.
     • Current going Left $\to$ Right was strong.
     • Current going Right $\to$ Left was weak (because the electrons got stuck at the "sponge" end).
   • Analogy: Imagine shouting from one side of a canyon. If you shout from the "safe" side, your voice carries clearly to the other side. If you shout from the "sponge" side, the wall absorbs your voice, and the other side hears almost nothing.

The "Exceptional Point": The Tipping Point

The paper also talks about something called an Exceptional Point (EP). Think of this as a tipping point or a phase transition.

• Below a certain magnetic strength, the electrons are mixed up, and the "skin" effect doesn't happen.
• Once the magnetic field crosses a specific threshold (the EP), the system suddenly flips. The electrons lock into their "spin" directions, the sponge starts working efficiently, and the "skin" effect kicks in.

The authors also solved a mystery: In these systems, the "tipping point" (where the effect starts) happens at a slightly different magnetic strength depending on whether the wire is infinitely long (theoretical) or a real, finite length. They used math to show exactly why this shift happens, proving that the length of the wire changes the rules slightly.

Why Does This Matter?

1. New Electronics: This could lead to electronic diodes or transistors that work based on these new quantum rules, allowing for one-way traffic of electricity without using traditional magnetic fields or diodes.
2. Probing the Invisible: The paper shows that by measuring "non-local" conductance (checking the far end of the wire), scientists can detect these strange quantum effects that are invisible if you only look at the local end.
3. Understanding the Future: As we move toward quantum computers, understanding how energy leaks and how electrons behave in "open" systems (systems that talk to the outside world) is crucial. This paper gives us a map for that behavior.

Summary in One Sentence

The paper shows that by connecting a special wire to a magnet, you can create a quantum "one-way street" where electrons pile up at one end, a phenomenon that can be detected by measuring how electricity flows differently depending on which end you push it from.

Technical Summary

1. Problem Statement

Open quantum systems described by non-Hermitian effective Hamiltonians exhibit unique phenomena, most notably the Non-Hermitian Skin Effect (NHSE), where eigenstates localize at system boundaries, and Exceptional Points (EPs), where eigenvalues and eigenvectors coalesce.

• The Challenge: While NHSE has been studied in photonic and chiral molecular systems, its manifestation in standard electronic transport remains unclear. Specifically, there is a lack of a concrete experimental proposal to detect NHSE in realistic electronic devices.
• The Gap: There is a disconnect between the topological classification of non-Hermitian systems (often derived under Periodic Boundary Conditions, PBC) and the behavior of finite systems (Open Boundary Conditions, OBC). Furthermore, the relationship between non-Hermitian eigenvectors and standard transport theory (conductance) needs clarification, particularly regarding non-reciprocity.

2. Methodology

The authors propose a realistic device setup and employ a combination of Green's function formalism, tight-binding models, and perturbation theory.

• Device Model:

  • A Rashba nanowire with strong spin-orbit coupling (SOC) is coupled to a ferromagnetic reservoir along its length.
  • The system is subjected to an axial magnetic field ($B$).
  • Geometry: The ferromagnet's spin polarization is aligned parallel to the SOC field and perpendicular to the magnetic field. This configuration creates a "helical" regime where spin and momentum are locked.
  • Coupling: The wire is weakly coupled to normal metallic leads (Left and Right) via gate-tunable barriers to measure transport.

• Theoretical Framework:

  • Effective Hamiltonian: The ferromagnetic contact is integrated out using the wide-band limit, resulting in a non-Hermitian effective Hamiltonian: $H_{eff} = H_0 + \Sigma^R$.
  • Self-Energy: The self-energy $\Sigma^R$ introduces spin-dependent dissipation: $\Sigma^R = -i(\gamma_0 + \gamma_y \sigma_y)$. Here, $\gamma_y$ represents spin-oriented dissipation.
  • Transport Calculation: Conductance ($G_{\alpha\beta} = dI_\alpha/dV_\beta$) is calculated using the Keldysh Green's function formalism. The conductance is expressed in terms of biorthogonal eigenstates (left $|\psi^l\rangle$ and right $|\psi^r\rangle$ eigenvectors), which differ in non-Hermitian systems.

3. Key Contributions

1. Experimental Proposal: The paper provides a concrete proposal to detect the NHSE in a semiconductor nanowire coupled to a ferromagnet, a system potentially realizable with current technology.
2. Transport Signature of NHSE: The authors establish a direct link between the NHSE and non-reciprocal non-local conductance. They demonstrate that while local conductance remains symmetric, non-local conductance becomes highly asymmetric depending on the direction of current flow.
3. Resolution of EP Shift: The paper explains a previously unexplained phenomenon: the shift of Exceptional Points (EPs) in parameter space when transitioning from PBC (bulk) to OBC (finite wire). They derive an analytical formula predicting this shift based on the wire length and mode index.

4. Key Results

A. Non-Reciprocal Transport

• Local Conductance ($G_{LL}, G_{RR}$): Remains symmetric ($G_{LL} = G_{RR}$) across all Zeeman energies. This is because local conductance probes the Local Density of States (LDOS), which remains symmetric even in the presence of NHSE.
• Non-Local Conductance ($G_{LR}, G_{RL}$): Exhibits pronounced non-reciprocity ($G_{LR} \gg G_{RL}$) in the helical regime (high magnetic field $B > \gamma_y$).
  • Mechanism: In the helical regime, right-movers have spins antiparallel to the ferromagnet (low dissipation), while left-movers have spins parallel (high dissipation).
  • NHSE Connection: The right eigenstates ($|\psi^r\rangle$) accumulate at one boundary (the Right contact), while left eigenstates ($|\psi^l\rangle$) accumulate at the opposite boundary. Non-local transport is maximized when the source and drain align with the localization of the right and left eigenstates, respectively.

B. Topological Phase Transition

• The system undergoes a non-Hermitian topological phase transition at $B = \gamma_y$.
• PBC (Bulk): The transition is marked by the closing of an imaginary gap and the coalescence of eigenvalues at an EP. The topological invariant (winding number) changes from 0 to 1.
• OBC (Finite Wire): The conductance peaks bifurcate, signaling the transition. However, the EPs for finite wires do not occur at the same $B$ as the bulk prediction.

C. Shift of Exceptional Points (EPs)

• Observation: In finite wires, the EPs occur at Zeeman energies $B > \gamma_y$, shifting away from the bulk critical point.
• Analytical Explanation: Using perturbation theory on the two-state subspace coalescing at the EP, the authors derive the condition for the EP in OBC:
  $$w_\nu^{OBC} = \text{sign}(B^2 |\kappa_\nu|^2 - \gamma_y^2)$$
  where $|\kappa_\nu|^2$ depends on the wire length $L$ and the mode index $\nu$. This explains why the shift is level-dependent and matches numerical simulations perfectly.

5. Significance

• New Probe for Non-Hermitian Physics: The work establishes non-local transport spectroscopy as a robust tool to probe NHSE and non-Hermitian topology in electronic systems, distinguishing it from Hermitian effects.
• Fundamental Understanding: It clarifies the relationship between non-Hermitian eigenvectors and transport, showing that non-reciprocity is a direct consequence of the biorthogonal nature of the eigenstates and their spatial localization.
• Resolution of Theoretical Discrepancies: The paper resolves the long-standing puzzle of why EPs shift in finite systems compared to bulk predictions, providing a perturbative framework applicable to other non-Hermitian systems.
• Robustness: The NHSE feature is topological (dependent on a point gap), implying it persists under disorder that does not close the gap, making it a viable target for experimental observation.

In summary, this paper bridges the gap between abstract non-Hermitian topology and measurable electronic transport, offering a clear experimental signature (non-reciprocal non-local conductance) for the Non-Hermitian Skin Effect in realistic nanowire devices.