⚛️ quantum physics

Pulsed two-photon scattering from a single atom in a waveguide with delay-modified temporal correlations

This paper theoretically investigates how delay-controlled temporal correlations in a bimodal two-photon pulse affect nonlinear scattering from a single atom chirally coupled to a waveguide, revealing striking differences in quantum correlations between uncorrelated and correlated photon pairs using matrix product states and frequency-dependent scattering theory.

Original authors: Matthew Kozma, Sofia Arranz Regidor, Stephen Hughes

Published 2026-03-24

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Matthew Kozma, Sofia Arranz Regidor, Stephen Hughes

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

The Big Picture: A Quantum Traffic Jam

Imagine you are trying to build a super-fast computer that uses light (photons) instead of electricity. To make this computer work, you need the light particles to "talk" to each other. In the real world, light beams usually pass right through each other without interacting (like two cars driving on parallel highways).

To make them interact, scientists use a tiny "traffic cop"—a single atom (or a quantum dot)—sitting in a narrow tunnel called a waveguide. This atom can grab a photon, get excited, and then spit it back out. This interaction creates a "non-linear" effect, which is the secret sauce for quantum computing.

The Problem: Usually, this interaction is too weak to be useful unless you have a massive amount of light. But quantum computers need to work with just a few photons.

The Solution: This paper asks a simple question: What happens if we send two photons at the atom, but we control exactly how they arrive in time?

The Setup: The "Two-Peak" Pulse

The researchers didn't just send two photons randomly. They created a special "pulse" of light that looks like a dumbbell or a double-hump camel. It has two distinct peaks (humps) of light separated by a gap.

They tested two different ways of loading these two photons into the pulse:

The "Two Separate Cars" Scenario (|1⟩|1⟩):
Imagine the pulse has two humps. In this scenario, they put one photon in the first hump and one photon in the second hump. The photons are like two separate cars driving one after the other. They don't know each other exists until the second car arrives.
- Analogy: You ring a doorbell once, wait a minute, and then ring it again. The person inside (the atom) reacts to the first ring, settles down, and then reacts to the second ring.
The "Shared Taxi" Scenario (|2⟩):
In this scenario, both photons are spread out over the entire pulse. They are "delocalized." It's as if the two photons are sharing a single taxi that has two seats, but the taxi itself is stretched out over both humps. The photons are indistinguishable; you can't tell which one is in the first hump and which is in the second.
- Analogy: You ring the doorbell with a special "double-ring" sound that covers both the first and second humps simultaneously. The person inside hears a complex, overlapping sound immediately.

The Experiments: What Happens When They Hit the Atom?

The researchers simulated what happens when these pulses hit the atom, looking at two main things:

Excitation: How "excited" does the atom get? (Does it jump up and down?)
Correlation: Do the photons come out in a synchronized pattern? (Do they stick together or scatter apart?)

1. The "Top-Hat" Pulse (The Blocky Experiment)

First, they used simple, block-shaped pulses (like a square wave).

Result for "Two Separate Cars": The first photon hits the atom, excites it, and leaves. The atom calms down. Then the second photon arrives and does the same thing. They act like two independent events. The atom never gets "overwhelmed."
Result for "Shared Taxi": Because both photons are spread out, they hit the atom at the same time (in a quantum sense). The atom gets excited immediately and starts behaving wildly. It can't just absorb one; it has to deal with two at once. This creates a "non-linear" effect where the output is totally different from the input.

2. The "Gaussian" Pulse (The Smooth Experiment)

Next, they used smooth, bell-shaped curves (more realistic for real lasers).

The "Sweet Spot": They found that if the two humps are too far apart, the atom treats them as separate events. If they are perfectly overlapping, they act like a single strong burst.
The Magic Middle: The most interesting thing happened when the humps were slightly separated.
- If the photons were "separate cars," the atom reacted linearly (boring).
- If the photons were "shared," even a tiny bit of separation caused the atom to get confused and excited in a complex way. The photons started "bunching" together (like birds flocking) in a pattern the researchers call "bird-like statistics."

The Key Discovery: Delocalization is the Switch

The most important finding is that how you arrange the photons matters more than how many you have.

If you keep the photons strictly separated in time (one here, one there), the system acts linearly (predictable, boring).
If you let the photons "delocalize" (spread out and overlap in time), the system instantly becomes non-linear (unpredictable, powerful).

It's like the difference between two people knocking on a door one second apart (the person inside opens the door twice) versus two people knocking at the exact same time (the person inside is startled and might do something unexpected).

Why Does This Matter?

This research shows that we don't need complex, expensive equipment to get strong quantum effects. We just need to tweak the timing of the light pulses.

By controlling the "delay" between the peaks of a light pulse, we can switch a quantum system from being a passive observer to an active, non-linear processor. This is a huge step forward for building:

Quantum Computers: Which need photons to interact to process data.
Quantum Sensors: Which need to detect tiny changes in light.
Secure Communication: Which relies on the unique correlations between photons.

Summary in a Nutshell

Think of the atom as a drummer and the photons as drumsticks.

Scenario A: You hit the drum once, wait, and hit it again. The drummer just plays two beats.
Scenario B: You hit the drum with two sticks that are slightly out of sync but overlapping. The drummer gets confused, the rhythm changes, and you get a complex, new sound.

This paper proves that by simply adjusting the "timing" of the drumsticks (the photons), we can force the drummer (the atom) to play a completely new, complex song that is essential for the future of quantum technology.

1. Problem Statement

Quantum nonlinearity is a fundamental requirement for advanced quantum technologies (computing, communication, sensing), yet it is often too weak to be exploited at the few-photon level. While single-photon interactions with a two-level system (TLS) in a waveguide are well-understood and largely linear, two-photon interactions introduce complex nonlinear phenomena due to saturation effects and the induction of photon-photon correlations.

The specific gap this paper addresses is the lack of detailed study regarding delay-modified temporal correlations in pulsed two-photon scattering. Specifically, the authors investigate how the localization of photons within a bimodal pulse envelope affects the system's dynamics. They compare two distinct input states:

Uncorrelated/Distinguishable State ( $|1\rangle|1\rangle$ ): Two single photons, each localized to a separate peak of the pulse envelope.
Correlated/Indistinguishable State ( $|2\rangle$ ): Two photons delocalized (but correlated) across the entire pulse envelope, sharing both peaks.

The study aims to quantify how the temporal separation ( $t_b$ ) between these pulse peaks influences population dynamics and photon correlation functions in a chiral waveguide-QED system.

2. Methodology

The authors employ two complementary theoretical frameworks to model a single TLS chirally coupled to a 1D waveguide:

Matrix Product States (MPS):
- Used for numerically exact simulations in a time-discretized basis.
- The system is modeled using time-bin noise operators.
- Input States: Constructed by applying creation operators to the vacuum. The $|1\rangle|1\rangle$ state is modeled as two independent chains of time bins, while the $|2\rangle$ state is a single chain with two photons distributed over the envelope.
- Limitation: The standard MPS implementation used here requires indistinguishable photons within a single Fock pulse, meaning the $|1\rangle|1\rangle$ case requires non-overlapping pulses ( $t_b \geq t_p$ ) to avoid representing them as a single indistinguishable state.
Frequency-Dependent Scattering Theory:
- Based on input-output theory and the two-photon scattering matrix ( $S$ -matrix).
- Solves for asymptotic output fields by transforming operators into the frequency domain.
- Advantage: Allows for continuous control over the spectral profile. The authors introduce a mixing parameter $\alpha \in [0, 0.5]$ to interpolate between the $|1\rangle|1\rangle$ $∣1 ⟩ ∣1 ⟩$ and $|2\rangle$ $∣2 ⟩$ states.
  - $\alpha = 0$ : Fully localized ( $|1\rangle|1\rangle$ ).
  - $\alpha = 0.5$ : Fully delocalized ( $|2\rangle$ ).
  - $0 < \alpha < 0.5$ : Intermediate states where photons are partially delocalized.
Pulse Envelopes:
- Top-hat pulses: Used for analytical clarity to define sharp boundaries between pulse activity and spontaneous emission.
- Gaussian pulses: Used for physical realism, allowing for the study of pulse overlap and interference.

3. Key Contributions

Comparative Dynamics: A rigorous comparison of scattering statistics between fully correlated ( $|2\rangle$ ) and uncorrelated ( $|1\rangle|1\rangle$ ) two-photon states, demonstrating that identical incident fluxes can yield radically different nonlinear responses.
Delay-Dependent Correlations: The identification of how the temporal separation ( $t_b$ ) between pulse peaks dictates the transition from linear to nonlinear scattering regimes.
Interpolation Framework: The development of a spectral mixing parameter ( $\alpha$ ) to continuously bridge the gap between distinguishable and indistinguishable photon inputs, revealing that any degree of delocalization immediately breaks the single-photon subspace restriction.
Validation: Cross-validation of results between MPS (time-domain) and Scattering Theory (frequency-domain), confirming consistency across different separation regimes.

4. Key Results

A. Population Dynamics

$|1\rangle|1\rangle$ (Localized): For non-overlapping pulses, the first pulse behaves linearly, exciting the TLS as a single photon. Nonlinear dynamics (stimulated emission) and non-zero second-order correlations ( $n_{TT}$ ) only occur upon the arrival of the second pulse.
$|2\rangle$ (Delocalized): Nonlinear dynamics occur immediately upon the arrival of the first pulse because the system is already in the two-photon subspace.
Effect of Separation ( $t_b$ ):
- Small $t_b$ : Strong bunching and complex interference.
- Large $t_b$ : The two peaks behave as independent excitation events.
- Intermediate $t_b$ : The most complex dynamics occur here, where the TLS is still saturated from the first pulse when the second arrives.

B. Second-Order Correlation Functions ( $G^{(2)}_{TT}$ )

$|1\rangle|1\rangle$ : $G^{(2)}_{TT}$ is zero until the second pulse arrives ( $t > t_b$ ). Once the second pulse interacts, distinct time periods emerge (initial pulse, break, second pulse, decay).
$|2\rangle$ : $G^{(2)}_{TT}$ $G_{T T}^{(2)}$ is non-zero immediately.
- Short $t_b$ : Exhibits "bird-like" bunching statistics (a known signature of two-photon scattering off a quantum dot).
- Long $t_b$ : Shows two distinct instances of the bird-like pattern corresponding to the two pulse peaks.
Interpolation ( $\alpha$ ): Even a small amount of delocalization ( $\alpha = 0.05$ ) results in non-zero $G^{(2)}_{TT}$ during the first pulse, signifying an immediate exit from the single-photon subspace.

C. Maximum TLS Excitation

Increasing the delocalization parameter $\alpha$ systematically reduces the maximum population achieved by the TLS.
This is attributed to the two-photon content inducing immediate stimulated emission, which depletes the population transfer to the excited state.
A global minimum in maximum population is observed when the separation $t_b$ is between $2\gamma^{-1}$ and $4\gamma^{-1}$ , where the TLS remains significantly saturated.

D. First-Order Correlations ( $G^{(1)}_{TT}$ )

In the limit of large separation (linear regime), the second-order correlation function is equivalent to the time-translated square of the first-order correlation function.
Nonlinear scattering contributions modify these features, curving the "zero lines" observed in the correlation maps.

5. Significance and Implications

Experimental Feasibility: The study demonstrates that these effects are accessible with current technologies, such as semiconductor quantum dot waveguides or circuit QED systems.
Control Mechanism: The results highlight temporal delay and photon localization as powerful control knobs for engineering quantum nonlinearities. By tuning the pulse separation and the degree of photon indistinguishability, one can switch between linear and highly nonlinear scattering regimes.
Methodological Insight: The paper validates the use of MPS for time-domain pulse analysis while acknowledging the flexibility of scattering theory for continuous spectral interpolation. It provides a roadmap for simulating more complex multi-photon states and entangled inputs.
Quantum Information: Understanding these correlations is crucial for designing quantum gates and logic operations that rely on photon-photon interactions mediated by a single emitter.

In conclusion, the paper establishes that the temporal structure of a two-photon pulse is as critical as its spectral content in determining the nonlinear response of a waveguide-QED system, offering new avenues for controlling quantum light-matter interactions.