Directional and correlated optical emission from a waveguide-engineered molecule with local control

This paper demonstrates that two quantum dots separated by 26 effective wavelengths in a bidirectional photonic crystal waveguide can be radiatively coupled to form an artificial molecule, enabling independent electrical control of directional and correlated optical emission through dispersive dipole-dipole interactions.

The Gist

Imagine you have two tiny, glowing fireflies (quantum dots) floating in a dark room. Normally, if you flick a switch to make them glow, they would just shine light in all directions, like a lightbulb. But what if you could make them shine only to the left, or only to the right, just by changing the timing of your flick?

That is exactly what this team of scientists achieved, but with light particles (photons) instead of fireflies, and using a high-tech "hallway" for light called a photonic crystal waveguide.

Here is the story of their experiment, broken down into simple concepts:

1. The Setup: A Long Hallway and Two Distant Dancers

The scientists built a microscopic hallway made of a special material (a photonic crystal). Inside this hallway, they placed two quantum dots (our "fireflies").

• The Distance: These two dots were placed 13 micrometers apart. To put that in perspective, that's like two people standing on opposite sides of a football field, but they are still holding hands through an invisible string.
• The String: That "invisible string" is the waveguide itself. Even though the dots are far apart, they are connected by the way light travels through the hallway. They can "talk" to each other through this light channel.

2. The Magic Trick: The "Artificial Molecule"

Usually, when two things are far apart, they act independently. But because these dots are connected by the waveguide, they start acting like a single unit, or what the scientists call an "artificial molecule."

Think of it like two dancers on a stage. If they dance perfectly in sync, they create a beautiful, unified performance. If they dance out of sync, they might cancel each other out.

• The Dispersive Interaction: The scientists found that the "string" between them wasn't just a simple connection; it had a twist. This twist (called a dispersive interaction) shifted the energy of their combined state, making them behave like a single, complex molecule rather than two separate dots.

3. The Control Knob: The "Phase" Switch

This is the coolest part. The scientists could control which way the light went by changing the timing (phase) of the laser pulses hitting the two dots.

• The Analogy: Imagine two people shouting into a long tunnel.
  • If they shout at the exact same time, the sound waves might cancel out on the left side and boost up on the right side.
  • If they shout slightly out of sync, the sound might cancel out on the right and boost on the left.
• The Result: By simply adjusting the "phase" (the timing) of the laser, the scientists could instantly switch the emission direction. They could make the light shoot only to the left or only to the right. It's like having a light switch that doesn't just turn the light on or off, but decides which room the light goes into.

4. The Traffic Police: Sorting Single Particles from Pairs

The experiment didn't just stop at directing light; it also sorted the type of light particles coming out.

• The Scenario: Imagine a busy train station. Usually, trains leave in random groups.
• The Discovery: The scientists found that under certain conditions, the "left exit" would only let out single passengers (single photons), while the "right exit" would only let out pairs of passengers (photon pairs) holding hands.
• Why it matters: This is huge for quantum computing. You need specific types of light particles to carry information. Being able to sort them automatically as they exit the system is like having a smart traffic cop that directs single cars to one lane and convoys to another.

5. The Full Inversion: The "High-Five"

Finally, the scientists managed to fully "flip" both dots at the exact same time.

• The Analogy: Imagine two people jumping up and down. If they jump together perfectly, they land in a way that creates a special, synchronized bounce.
• The Result: When they did this, they observed that the two dots emitted correlated pairs of photons. It was as if the first photon "told" the second photon, "Hey, I'm going this way, you should follow me!" This creates a stream of linked light particles, which is a building block for future quantum networks.

Why Does This Matter?

This work is a major step toward the Quantum Internet.

• The Problem: Currently, building quantum networks is hard because we can't easily control where quantum information goes. It's like trying to send a letter but not knowing which mailbox it will land in.
• The Solution: This experiment shows we can build "waveguide-engineered molecules" that act as smart routers. We can tell a quantum bit of information exactly which path to take.
• The Future: The scientists showed that this system can be scaled up. They proved that with a few more "trenches" (cuts in the material) to control more dots, we could eventually build a network with many emitters, creating complex quantum computers and unbreakable communication lines.

In a nutshell: The scientists took two distant quantum dots, tied them together with a light highway, and taught them to dance in a way that lets them choose exactly which direction to shine their light, effectively creating a microscopic, programmable traffic system for the future of quantum technology.

Technical Summary

1. Problem Statement

Quantum networks require efficient routing of quantum information between distant nodes. While waveguide Quantum Electrodynamics (QED) offers a natural framework for generating and manipulating propagating photons, typical waveguide structures emit photons symmetrically to both output ports. Breaking this symmetry to achieve chiral quantum optics (directional emission) is crucial for scalable quantum networks.

Existing methods for directional emission often rely on:

• Chiral waveguides: Breaking reflection symmetry in the waveguide structure itself (e.g., reflection-symmetry broken photonic crystals).
• Microwave regimes: Realizing "artificial molecules" with superconducting qubits, which is difficult to translate to the optical domain.
• Single emitters: Using spin-orbit coupling or specific atomic transitions, which limits scalability.

The Gap: There has been no demonstration in the optical domain of controlling the emission direction of spatially separated, independently tunable quantum emitters coupled via a non-chiral (bidirectional) waveguide. The challenge lies in achieving radiative coupling over large distances (many wavelengths) while maintaining independent electrical control to tune the emitters into resonance.

2. Methodology

The authors developed a platform based on InGaAs quantum dots (QDs) embedded in a bidirectional photonic crystal waveguide (PCW) patterned in a suspended GaAs membrane.

• Device Architecture:

  • Independent Tuning: A shallow trench etched through the p-doped layer of the p-i-n diode structure electrically isolates the left and right halves of the waveguide. This allows independent application of DC Stark shifts to tune the emission frequency of QDs in each half.
  • Emitter Separation: Two QDs are selected with a separation of 13 µm (approximately 26 effective wavelengths at the band edge).
  • Optical Control: A Spatial Light Modulator (SLM) is used to create two coherent laser beams from free space, allowing independent control over the driving strength, position, and relative phase ($\theta_d$) of the excitation fields.

• Experimental Regimes:

  1. Resonance Tuning: Using DC bias voltages to bring the two QDs into spectral resonance.
  2. Pulsed Collective Excitation: Using the SLM to drive both QDs simultaneously with controlled phase to study single-photon emission directionality.
  3. Continuous Driving: Studying photon statistics ($g^{(2)}$) under weak continuous drive to observe correlations.
  4. Full Inversion: Using $\pi$-pulses to fully invert both emitters (creating a $|ee\rangle$ state) to generate correlated photon pairs.

3. Key Contributions

• Waveguide-Engineered Molecule: The authors demonstrate that two spatially separated QDs (13 µm apart) coupled via a common guided mode form an "artificial molecule." The coupling is not purely dissipative but possesses a significant dispersive component ($J_{12}$).
• Phase-Controlled Directionality: They show that the emission direction can be switched from left to right (and vice versa) solely by manipulating the relative phase of the driving fields, without changing the waveguide geometry.
• Directional Photon Statistics: The work reveals a novel regime where single photons are emitted predominantly to one port while photon pairs are emitted to the other, depending on the driving conditions.
• Correlated Emission: They demonstrate the generation of temporally correlated photon pairs with specific directional properties (emergent chirality) resulting from the cascaded decay of the doubly excited state.

4. Key Results

A. Radiative Coupling and Dispersive Interaction

• Transmission Spectroscopy: When both QDs are tuned to resonance, the transmission dip is deeper than the product of individual dips, confirming radiative coupling.
• Asymmetry: The transmission dip is asymmetric with respect to detuning, indicating a dispersive coupling component ($J_{12} \propto \sin \phi$) rather than purely dissipative coupling. The coupling phase is determined to be $\phi \approx 0.8\pi$.
• Molecular States: The system forms collective states $|\pm\rangle = (|eg\rangle \pm e^{i\phi}|ge\rangle)/\sqrt{2}$. The dispersive interaction shifts the energy levels of these states, effectively creating a molecule.

B. Controllable Directional Emission

• Phase Switching: By varying the relative driving phase $\theta_d$, the emission direction is switched.
  • When $\theta_d = \pi - \phi$, destructive interference cancels the left-going field ($E_L = 0$).
  • When $\theta_d = \pi + \phi$, the right-going field is suppressed ($E_R = 0$).
• Efficiency: The directionality ratio ($I_L / (I_L + I_R)$) can be tuned from near 0 to near 1, demonstrating on-demand routing of single photons.

C. Directional Photon Statistics ($g^{(2)}$)

• Single-Photon vs. Pair Emission: Under continuous weak driving of only one QD:
  • Left Port (Driven side): Shows strong antibunching ($g^{(2)}(0) \approx 0.31$), indicating single-photon emission.
  • Right Port (Undriven side): Shows an "antidip" with bunching ($g^{(2)}(0) \approx 0.98$), indicating the emission of photon pairs.
• Mechanism: The driven QD acts as a source; the undriven QD acts as a saturable mirror. The system preferentially emits single photons to the driven side and pairs to the opposite side due to the interference of the collective states.

D. Correlated Photon Pairs from Full Inversion

• Cascaded Decay: When both QDs are fully inverted to the $|ee\rangle$ state, they decay via a two-step process.
• Temporal Correlation: Time-resolved measurements show that two photons are emitted together (temporally correlated).
• Emergent Chirality: The emission exhibits a contrast between same-port ($g^{(2)}_{LL}, g^{(2)}_{RR}$) and cross-port ($g^{(2)}_{LR}, g^{(2)}_{RL}$) correlations, a signature of spontaneously broken mirror symmetry.

5. Significance and Outlook

• Novel Platform for Chiral QED: This work establishes a route to chiral quantum optics using non-chiral waveguides. The chirality is engineered dynamically via the relative phase of the drive and the dispersive coupling of the emitters, offering a more flexible alternative to static chiral waveguide designs.
• Scalability: The authors provide a feasibility analysis (Appendix G) showing that scaling to sets of 3–4 resonant QDs is achievable with current technology (using multiple trenches). Scaling to 10+ emitters is predicted to be possible with moderate improvements in QD density, spectral homogeneity, and tuning range.
• Quantum Networking: The ability to independently tune, address, and route photons from multiple emitters is a critical step toward:
  • Deterministic generation of spin-spin entanglement.
  • Implementation of quantum gates.
  • Generation of multi-photon cluster states for photonic quantum computing.
• Fundamental Physics: The observation of dispersive coupling over 26 wavelengths in the optical domain opens new avenues for studying many-body physics and collective radiance in waveguide QED.

In summary, this paper demonstrates a versatile, electrically tunable platform where distant quantum emitters form a controllable "artificial molecule," enabling dynamic switching of emission direction and the generation of complex correlated photonic states.