🔬 atomic physics

A single atom emitting resonance fluorescence into a coherent beam, and its use for non-destructive atom thermometry

This paper demonstrates a non-destructive, time-resolved atom thermometry technique by analyzing the interference between a single $^{87}$ Rb atom's resonance fluorescence and a coherent beam, achieving 4% temperature uncertainty at $\sim 30 \mu\mathrm{K}$ with 200 $\mu$ s resolution.

Original authors: Tomáš Lamich, Laura Zarraoa, Sondos Elsehimy, Morgan W. Mitchell, Romain Veyron

Published 2026-08-03

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Tomáš Lamich, Laura Zarraoa, Sondos Elsehimy, Morgan W. Mitchell, Romain Veyron

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 the smallest building blocks of matter—atoms—are not just invisible specks, but tiny, glowing actors on a stage. In the theater of quantum physics, scientists have long been trying to watch these actors perform without accidentally knocking them off the stage. Usually, to measure how fast an atom is jiggling (its temperature), you have to let it go and watch it fly away, which destroys the experiment. It's like trying to measure the speed of a hummingbird by catching it in a net; you get the answer, but the bird is gone.

To understand the new trick described in this paper, we need two main ideas. First, think of a "coherent beam" as a perfectly synchronized army of light waves marching in step, like a choir singing a single, pure note. Second, imagine an atom as a tiny mirror that can also sing. When this atom hears the choir, it starts to vibrate and sing back, creating its own little wave of light. The big question scientists have been asking is: What happens when the atom's song mixes with the choir's song? Does it make the sound louder, quieter, or something in between? This isn't just about making noise; it's about using that mixing to measure the atom's temperature without ever touching it, keeping the atom safe and sound for future quantum computers and super-precise clocks.

The Paper: A Single Atom as a Tiny Thermostat

In this study, a team of researchers at ICFO in Spain decided to play a game of "light interference" with a single atom. They trapped a lone Rubidium-87 atom in a cage made of laser light, known as a far-off-resonance trap (FORT). Think of this trap as a bowl of invisible jelly that holds the atom in place. Once the atom was cozy in its bowl, the scientists shined two beams of light on it. One was a strong "pump" beam that made the atom vibrate and glow (resonance fluorescence), and the other was a weak "probe" beam that acted like a gentle tap to see how the atom was reacting.

Here is the magic trick: The light bouncing off the atom (the fluorescence) and the weak probe beam traveling through the trap met up at the detector. Because light acts like a wave, these two beams interfered with each other. Sometimes, their peaks lined up and made the signal brighter (constructive interference); other times, a peak met a trough and canceled each other out, making the signal dimmer (destructive interference). It's like two people pushing a swing: if they push at the same time, it goes high; if one pushes while the other pulls back, it stops.

The researchers found that how much the signal flickered between bright and dark depended entirely on where the atom was sitting inside its light-bowl. If the atom was frozen in the very center, the interference was sharp and clear. But if the atom was jittering around (which means it was hot), it moved through different parts of the light waves, blurring the interference pattern. By watching how "fuzzy" the interference was, the scientists could figure out exactly how hot the atom was, all without destroying it.

The Findings: A New Way to Measure Heat

The team tested this method on 1,200 different atoms. They would cool an atom down, shine the lights on it for a tiny slice of time (80 milliseconds total, broken into 1-millisecond chunks), and count the photons that arrived. They repeated this 80 times for each atom.

What they discovered was that this "interference visibility" acts as a super-sensitive thermometer. When they compared their new method to the old, destructive way of measuring temperature (called release-and-recapture, where the atom is let go and caught again), the results matched up perfectly. The new method showed that the atom's temperature rose from about 18 microkelvin (very cold!) to 133 microkelvin during the experiment, just as the old method predicted.

The best part? This new thermometer is non-destructive and incredibly fast. The researchers could measure the temperature with a time resolution of about 200 microseconds (that's 0.0002 seconds) and get an uncertainty of only 4%. In comparison, the old method took longer and had a much bigger margin of error (around 10%), plus it ruined the atom in the process.

The paper confirms that by analyzing the statistics of the photons—essentially counting how many light particles arrived and when—they could infer the atom's position and temperature. They ruled out the idea that this was just random noise; the pattern was clear and consistent with the laws of physics. While the method works best when the atom is relatively cool (below about 70 microkelvin for the most stable readings), it opens the door to watching atoms in real-time. This could help scientists build better quantum computers by letting them "feel" the temperature of their qubits (quantum bits) without breaking them, perhaps even using this light-mixing trick to cool the atoms down further in the future.

Technical Summary: Single-Atom Resonance Fluorescence Interference for Non-Destructive Thermometry

Problem Statement
Accurate measurement of the center-of-mass temperature of a single trapped atom is a critical requirement for quantum computation, simulation, and networking, as temperature directly impacts gate fidelity and memory storage times. However, measuring single-atom temperature is inherently challenging due to weak signals and the statistical nature of photon detection. Existing non-destructive methods, such as resolved sideband cooling scattering or self-interference of resonance fluorescence, have limitations. Conversely, destructive methods like release-and-recapture (R&R) or time-of-flight (TOF) destroy the atomic state, preventing continuous monitoring or repeated measurements on the same atom. There is a need for a sensitive, time-resolved, non-destructive thermometry technique that can operate on the microsecond timescale.

Methodology
The authors investigate a light-matter interaction scenario involving a single neutral $^{87}\text{Rb}$ atom trapped in a far-off-resonance optical dipole trap (FORT). The experimental setup utilizes a specific geometry where the atom is simultaneously illuminated by two coherent beams:

A strong, circularly polarized "pump" beam propagating along the Z-axis, resonant with a closed transition ( $F, m_F = 2, -2 \to F, m_F = 3, -3$ ), which drives the atom to scatter photons (resonance fluorescence).
A weak, linearly polarized "probe" beam propagating along the Y-axis, sharing the same frequency as the pump.

The pump induces resonance fluorescence, while the probe beam interferes with the light scattered by the atom into the probe mode. The collected light (probe + scattered fluorescence) is directed into a single-mode fiber and detected by a single-photon counting module (SPCM).

The experiment involves loading atoms into the FORT, cooling them via polarization gradient cooling (PGC), and then subjecting them to 80 cycles of 1 ms exposure to the pump and probe beams. Photon arrival times are time-tagged. The researchers analyze the photon count statistics within 250 $\mu$ s time bins to observe how the interference pattern evolves as the atom heats up during the exposure.

Theoretical modeling extends the work of Goncalves et al. [Phys. Rev. A 104, 013724] to include fluctuating experimental conditions and atomic motion. The model treats the atom as a two-level system driven by the pump and probe fields. The detected photon flux is derived from the first-order correlation function, showing that the mean collected photon number depends on the interference term $\cos \phi$ , where $\phi$ is the relative phase between the pump and probe at the atom's position. This phase includes a component dependent on the atom's center-of-mass displacement ( $\phi_{cm}$ ). As the atom's temperature increases, its spatial distribution widens, leading to an averaging of the phase and a reduction in interference visibility.

Key Contributions

Observation of First-Order Interference: The paper experimentally confirms the prediction that a single atom undergoing resonance fluorescence can interfere with a coherent probe beam, resulting in either amplification or de-amplification (destruction) of the beam flux depending on the relative phase.
Non-Destructive Thermometry Model: The authors develop a statistical model for the photon counting distribution that accounts for atomic motion (thermal broadening) and experimental drifts. They derive a relationship between the interference visibility and the atom's temperature.
Time-Resolved Measurement: By analyzing photon counts in short time bins, the method allows for the reconstruction of temperature evolution over time without destroying the atom.

Results

Interference Visibility: The experiment demonstrates clear oscillations in photon counts as the relative phase drifts, confirming the interference effect. The visibility of this interference is shown to be highly sensitive to the atom's localization within the trap.
Temperature Correlation: As the atom heats up during the 1 ms exposure (from an initial $\sim 18 \, \mu\text{K}$ to $\sim 133 \, \mu\text{K}$ as measured by destructive R&R), the interference visibility decreases. This reduction is attributed to the atom exploring a larger phase space, which averages out the interference term.
Quantitative Performance: Using 1200 atoms with an integration time of 80 ms per atom (80 cycles of 1 ms), the method achieves a temperature uncertainty of 4% for temperatures around 30 $\mu\text{K}$ . The time resolution is approximately 200 $\mu\text{s}$ .
Comparison: The new method's temperature estimates agree well with those obtained from the destructive R&R method but offer significantly better relative uncertainty (4% vs. $\sim$ 10% for R&R) and, crucially, do not destroy the atomic state. The technique is robust against variations in the ratio of probe to scattered light brightness ( $N_{pr}/N_{sc}$ ).

Significance and Claims
The paper claims to demonstrate a viable method for non-destructive, time-resolved thermometry of a single trapped atom. By fitting the photon count distributions to their derived model, the authors can infer the center-of-mass localization and thus the temperature of the atom in real-time.

The authors state that this method could be implemented in various trapped neutral atom and ion systems. They suggest potential applications in optimizing cooling methods, such as feedback cooling, by providing continuous temperature feedback. They also note that the interference signal could be utilized to stabilize optical paths to a single emitter. While the current work operates in a regime with similar fluxes of probe and scattered photons, the authors acknowledge that the method could theoretically be extended to high-flux probe regimes, resembling homodyne detection of the resonance fluorescence field. The work validates the theoretical predictions of Goncalves et al. regarding the statistical properties of this specific light-matter interaction scenario.

More like this