🔬 optics

Phase-controlled transport of Floquet-driven compact topological photonic states

This paper demonstrates that Floquet driving can overcome the inherent immobility of compact flat-band states in photonic lattices by inducing a phase-controlled, chiral transport mechanism governed by topological winding numbers, which is experimentally verified using femtosecond laser-written waveguides.

Original authors: Gabriel Caceres-Aravena, Paloma Vildoso, Helena Drüeke, Rodrigo A. Vicencio

Published 2026-07-14

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Gabriel Caceres-Aravena, Paloma Vildoso, Helena Drüeke, Rodrigo A. Vicencio

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a world where light is usually stuck in place, like a marble trapped in a perfectly smooth, flat bowl. In the physics of light, these "flat bowl" spots are called flat bands. Normally, if you drop a photon (a particle of light) into one of these spots, it just sits there. It can't move, it can't travel, and it certainly can't go anywhere. It's a total traffic jam for light, known as the "Aharonov-Bohm caging effect." For a long time, scientists thought that once light was trapped in these compact, zero-tail states, it was impossible to get it to move without using messy, nonlinear tricks.

But this paper says: Not so fast.

The researchers, working with a team from Germany and Chile, found a way to make this trapped light dance. They didn't just push it; they taught it a specific rhythm.

The Magic Trick: The Light Switch

Think of their setup as a long, narrow hallway made of glass waveguides (tiny tunnels for light). Inside this hallway, they built a special pattern called a "diamond lattice." In this pattern, the light is supposed to be stuck.

To get the light moving, they used a technique called Floquet driving. Imagine you are trying to push a heavy swing. If you push it randomly, it just wobbles. But if you push it at the exact right moment in its swing, it goes higher and higher.

In this experiment, the "push" is a switch. The scientists built their glass hallway in segments. Every 16.1 mm along the path, they physically swapped the type of tunnel in the center of the pattern.

Segment 1: The center has a "Type S" tunnel.
Segment 2: The center has a "Type P" tunnel.
Segment 3: Back to "Type S."
Segment 4: Back to "Type P."

They did this by using a super-fast laser to write these tunnels into a glass wafer that was 70 mm long. By alternating these "masks" (the S and P tunnel types), they created a periodic rhythm that the light could lock onto.

The Directional Dance

Here is the coolest part: the direction the light travels depends entirely on how you start the dance.

The "In-Phase" Start: If you shine light into the starting tunnels so that the waves are perfectly synchronized (like two people clapping at the exact same time), the light packet jumps to the left.
The "Out-of-Phase" Start: If you shine the light so the waves are opposite (one claps while the other stays silent), the light packet jumps to the right.

It's like a chiral pair of dancers. One version of the dance moves left, and its mirror-image partner moves right. The paper confirms that these traveling states are "chiral," meaning their direction is locked to their internal phase, carrying a topological number (a winding number) of +1 for right-moving and -1 for left-moving.

The Proof: Watching the Light Jump

The team didn't just simulate this on a computer; they actually built it. They used a femtosecond laser to write the waveguides into a borosilicate glass wafer. They shot a laser beam at 640 nm (a specific shade of red-orange) into the start of the glass.

They watched what happened as the light traveled through the glass.

After passing through one segment (16.1 mm), the light had moved one step.
After two segments, it moved two steps.
After three and four segments, it kept moving, step by step, exactly as predicted.

They measured the light's movement and found it was incredibly precise. For the first three segments, the light stayed almost perfectly compact, with only about 3% of the energy "leaking" out or getting lost. This is a big deal because, for years, people thought you needed complex, nonlinear materials to make compact light move. This paper shows you can do it in a simple, linear system just by switching the geometry.

However, the paper is careful to note that as the light traveled further (through the fourth segment), the "leakage" increased to about 7%. This isn't a failure; it's just the reality of building something so precise. Tiny imperfections in the glass and the fact that the light traveled a longer distance (70 mm total) caused a few more photons to wander off.

The Bottom Line

The researchers successfully demonstrated that you can take light that is naturally trapped and make it zip along a straight line, purely by changing the shape of the path it travels. They proved that by tuning the relative phase of the input light, you can control whether it goes left or right.

This isn't a magic wand that solves all energy problems yet, but it is a solid, measured proof that "flat band" light doesn't have to stay still. It showed that with the right rhythm—swapping the tunnels every 16.1 mm—you can turn a stationary light trap into a controlled, directional highway.

Technical Summary: Phase-controlled transport of Floquet-driven compact topological photonic states

Problem Statement
The localization, transport, and manipulation of energy are fundamental challenges in physics. In photonic lattices, Flat Band (FB) systems offer a solution for perfect energy localization due to destructive interference, resulting in compact states with zero tails. However, a significant limitation exists: in linear environments, these compact FB states are generally immobile. Their spatial confinement prevents transport, as they remain localized unless nonlinear effects are introduced. The authors identify the need for a mechanism to manipulate bulk excitations in a controlled manner without inducing spatial dispersion, specifically within a linear, all-FB lattice system.

Methodology
The authors propose and experimentally implement a Floquet engineering protocol to induce transport in an all-FB photonic lattice. The methodology involves the following components:

Lattice Design: The system utilizes a diamond (rhombic) lattice incorporating multi-orbital interactions between single-mode (S) and second-order mode (P) waveguides. This configuration induces a synthetic magnetic flux of $\pi$ , creating an Aharonov-Bohm (AB) caging effect that results in an all-FB spectrum with eigenvalues $\beta = \{0, \pm 2t\}$ .
Floquet Driving Protocol: To overcome the immobility of FB states, the Hamiltonian is modulated along the propagation coordinate ( $z$ ). The system consists of a periodic concatenation of two distinct lattice masks (Mask $a$ and Mask $b$ ). These masks differ only by the exchange of the central waveguides (swapping S and P sites) every half-period $Z_\pi = \pi/(2t)$ .
Theoretical Framework: The dynamics are analyzed using tight-binding approximations and Floquet theory. The authors calculate the Floquet quasienergies and the stroboscopic evolution operator. They determine the topological classification of the system, specifically looking for chiral symmetry and winding numbers ( $\nu$ ).
Experimental Implementation: The lattice was fabricated on a borosilicate glass wafer ( $L=70$ mm) using femtosecond laser writing. The writing power was tuned to create S-only and multimode P waveguides, ensuring the propagation constants of S and P modes matched at specific lattice positions to induce the required inter-orbital coupling. The protocol involved alternating the central waveguides between S and P types at intervals of $16.1$ mm. Input conditions were generated using spatial light modulators to create specific in-phase and out-of-phase excitations at $640$ nm.

Key Contributions and Results

Controlled Transport of Compact States: The primary result is the demonstration of directional transport of compact FB states in a purely linear system. By modulating the Hamiltonian, the authors transformed the static caging effect into a moving state.
Phase-Dependent Directionality: The direction of propagation is strictly determined by the relative phase of the input condition.
- In-phase excitation of specific vertical waveguides results in leftward propagation.
- Out-of-phase excitation of the same waveguides results in rightward propagation.
- This behavior is attributed to constructive and destructive interference mechanisms at the connector sites (S and P sites) which are swapped during the Floquet cycle.
Topological Characterization: The system is classified as an AIII Floquet topological insulator. The traveling states exist in chiral pairs with a topological invariant (winding number) of $\nu = \pm 1$ $ν = \pm 1$ .
- $\nu = 1$ corresponds to right-traveling states.
- $\nu = -1$ corresponds to left-traveling states.
- The system possesses chiral symmetry, protected by a chiral operator derived from the static Hamiltonians and the exchange operator of the masks.
Experimental Validation: The experiment successfully demonstrated discrete transport over four concatenated masks.
- The wave packet displacement was measured to be approximately 1, 2, 3, and 4 lattice units after each mask segment, confirming the stroboscopic quantum walk-like effect.
- Leakage was quantified using the participation ratio. For up to three masks, the average leakage was approximately 3%. With four masks, leakage increased to roughly 7%, attributed to cumulative fabrication defects and imperfect mode conversion during the waveguide swaps.

Significance and Claims
The paper claims to provide a mechanism for the controlled transport of compact photonic states without relying on nonlinear phenomena, which were previously thought necessary for such transport in discrete systems. By combining multi-orbital interactions with Floquet engineering, the authors demonstrate that the mobility of localized states can be fully controlled by the input phase and position.

The work extends the frontiers of photonic transport by achieving discrete, directional movement using only evanescent coupling, avoiding the extra losses associated with bending waveguides. The authors assert that their results confirm the Floquet-engineered system effectively drives the movement of FB states, with directionality governed by interference at connector sites. The experimental observation validates the theoretical prediction that compact states, typically immobile in static FB lattices, can be made to travel as topological chiral pairs in a driven linear environment.

The Magic Trick: The Light Switch

The Directional Dance

The Proof: Watching the Light Jump

The Bottom Line

More like this