In situ subwavelength microscopy of ultracold atoms using dressed excited states

This paper presents a novel method for subwavelength imaging of ultracold atoms using laser-driven excited state interactions to engineer ground state population transfer, demonstrating both strong and weak imaging regimes to achieve resolutions down to 30 nm while providing a general theoretical framework for their validity.

The Gist

Imagine you are trying to take a photo of a tiny, invisible ant sitting on a table. But there's a catch: your camera lens is too blurry to see anything smaller than a grain of sand. This is the "diffraction limit," a fundamental rule of physics that says you can't see details smaller than half the wavelength of the light you're using. For a long time, scientists thought this was a hard wall they couldn't break when looking at ultracold atoms.

This paper describes a clever new trick to break that wall. The researchers didn't just build a better lens; they taught the atoms to "paint" themselves with a resolution far finer than the light itself.

Here is the story of how they did it, using simple analogies.

The Problem: The Blurry Flashlight

Usually, to take a picture of atoms, scientists shine a light on them. If the atoms absorb the light, they get excited and glow, allowing the camera to see them. But because light acts like a wave, it creates a "fuzzy" spot. You can't distinguish two atoms if they are closer together than the width of that fuzzy spot (about 390 nanometers in this experiment).

The Solution: The "Dressed" State

The researchers used a special technique involving three levels of energy in the atoms (think of them as three floors in a building: Ground Floor, First Floor, and Second Floor).

1. The Setup: They have atoms sitting on the Ground Floor. They want to take a picture of them.
2. The Trap: They shine a very specific laser (1529 nm) that creates a "landscape" of energy hills and valleys. This laser doesn't just sit there; it "dresses" the atoms on the First Floor, changing their energy depending on exactly where they are standing.
3. The Trigger: Then, they use a second laser (the "repumper") to try to push atoms from the Ground Floor up to the Second Floor (where the camera can see them).

The Magic Trick:
Because of the "dressed" landscape created by the first laser, the second laser only works in a tiny, tiny slice of space. It's like having a key that only opens a door if you are standing exactly 30 nanometers away from a specific point. Everywhere else, the key doesn't fit.

By tuning this key perfectly, they can force atoms to jump to the "visible" floor only in a strip that is 30 nanometers wide. This is much, much smaller than the blurry spot of their camera lens.

Two Ways to Do It: The Sprint vs. The Stroll

The paper explores two different ways to use this trick, which the authors call "Strong Imaging" and "Weak Imaging."

1. The Strong Imaging Regime (The Sprint)

Imagine you have a crowd of people in a dark room, and you want to highlight only the people standing in a tiny circle.

• How it works: You turn on a super-bright spotlight for a split second (nanoseconds). It's so bright and fast that the people don't have time to move or react before the light hits them.
• The Result: You get a sharp, high-contrast image of that tiny circle.
• The Analogy: It's like taking a photo with a camera flash that is so fast it freezes a hummingbird's wings. The researchers used this method on a cloud of warm atoms and achieved a resolution of 100 nanometers.

2. The Weak Imaging Regime (The Stroll)

Now, imagine the same crowd, but you are very gentle.

• How it works: You use a dimmer light for a longer time. You are so gentle that the atoms don't get jostled or pushed around; they stay perfectly still in their original formation.
• The Result: Even though the light is weak, because you are so careful, you can pick out a single atom from a very tight group without disturbing the group's shape.
• The Analogy: It's like gently picking a single grape from a bunch without squishing the others. This is the "counter-intuitive" part: usually, we think we need to be fast to get a good picture, but here, being slow and gentle actually gave them the best results.
• The Result: They used this on a super-cold, tightly packed group of atoms (a Bose-Einstein Condensate) and resolved a wave pattern that was only 45 nanometers wide.

Why This Matters

Think of this like upgrading from a standard-definition TV to 8K, but for the quantum world.

• Before: Scientists could see "blobs" of atoms.
• Now: They can see the individual "pixels" of quantum matter, even when those pixels are smaller than the wavelength of light used to see them.

This is crucial for building quantum computers and simulating complex materials. Just as a biologist needs a microscope to see cells, a quantum physicist needs this "super-microscope" to see how atoms interact in tiny, engineered structures.

The Takeaway

The researchers didn't invent a better lens. Instead, they invented a new way to "ask" the atoms where they are. By using a clever combination of lasers to create a "selective door" that only opens for atoms in a specific, tiny spot, they bypassed the laws of blurry light. They proved that whether you sprint (strong imaging) or stroll (weak imaging), you can see the invisible world with incredible clarity.

Technical Summary

1. Problem Statement

Quantum gas microscopes are essential for simulating quantum many-body systems, but they are fundamentally limited by the diffraction limit of light ($\lambda/2$). While techniques like stroboscopic imaging or near-field lattices have been proposed to create subwavelength potentials, imaging these structures with subwavelength resolution remains a challenge.

• The Core Challenge: Standard imaging perturbs the system. To achieve subwavelength resolution, one must transfer atoms from a "dark" state to a "bright" state within a spatial region much smaller than the wavelength. However, this transfer induces dynamics (momentum kicks, wavefunction spreading) that can blur the image or alter the state being measured.
• Existing Limitations: Previous super-resolution methods (e.g., Refs. [15, 16]) relied on the "strong imaging" regime (rapid transfer) to freeze dynamics. There was a lack of a general theoretical framework to define the validity of different imaging regimes and to explore alternative approaches, such as "weak imaging," which might offer robustness with fewer timing constraints.

2. Methodology

The authors propose a novel imaging technique based on dressed state engineering in a three-level system ($|1\rangle, |2\rangle, |2'\rangle$) using $^{87}\text{Rb}$ atoms.

A. Physical Mechanism

1. Dressed Excited State: A 1529 nm laser lattice drives the transition between two excited states ($5^2P_{3/2}$ and $4^2D_{5/2}$). This creates a spatially dependent light shift (AC Stark shift) on the excited state $|2'\rangle$.
2. Spatial Selectivity: A repumper laser (780 nm) is tuned to couple the ground state $|1\rangle$ to the dressed excited state $|2'\rangle$. Because the energy of $|2'\rangle$ varies spatially (cosine profile), the repumper is only resonant at specific positions (slices) where the detuning matches the light shift.
3. Population Transfer: Atoms in the resonant slice are transferred from the dark ground state $|1\rangle$ to the bright ground state $|2\rangle$ via spontaneous emission cycles.
4. Imaging: After the transfer, the population in $|2\rangle$ is imaged using a cycling transition, while atoms remaining in $|1\rangle$ are ignored.

B. Theoretical Framework

The authors derive a general formalism using a Schrödinger equation with a non-Hermitian loss term to describe the spatio-temporal dynamics:
$$i\hbar \frac{d|\Psi(x, t)\rangle}{dt} = \hat{H}_0 |\Psi(x, t)\rangle - i\hbar \frac{\kappa(x)}{2} |\Psi(x, t)\rangle$$
where $\kappa(x)$ is the spatially dependent transfer rate. This formalism defines two distinct regimes:

1. Strong Imaging Regime (Diabatic): High transfer rate ($\kappa_0$) applied for a very short time ($t_{tr}$). The wavefunction is transferred before it has time to evolve spatially. Validity criterion: $S \gg 1$ (ratio of pumping strength to localization recoil).
2. Weak Imaging Regime (Adiabatic): Low transfer rate applied for a long time. The system evolves adiabatically, remaining in the ground state of the harmonic trap throughout the process. Validity criterion: $W \ll 1$ (adiabaticity parameter).

3. Key Contributions

• Dual-Regime Demonstration: The paper experimentally demonstrates that subwavelength resolution can be achieved in both the strong (diabatic) and weak (adiabatic) imaging regimes.
• General Validity Criteria: The authors derive analytical criteria ($S$ and $W$) that define the boundaries of validity for both regimes, applicable to other subwavelength methods.
• Robustness of Weak Imaging: Contrary to intuition, the weak imaging regime is shown to be a robust solution for resolving extremely narrow wavefunctions, as it avoids the rapid momentum kicks associated with strong imaging, provided the transfer is slow enough to be adiabatic.
• Quantitative Agreement: The experimental results show quantitative agreement with the analytical model without adjustable parameters, validating the theoretical formalism.

4. Experimental Results

A. Strong Imaging Regime (Thermal Gas)

• Setup: A thermal cloud of $^{87}\text{Rb}$ was imaged using a 1529 nm lattice with a period of 8.3 $\mu$m.
• Observation: By varying the light shift amplitude ($U_{5P,0}$) and repumper saturation, the authors measured the number of transferred atoms.
• Resolution: They achieved a Full Width at Half Maximum (FWHM) of the transferred slice of 100 nm, significantly below the diffraction limit ($\lambda/2 \approx 390$ nm).
• Validation: The experimental atom counts matched the theoretical model (Eq. 15) perfectly when correcting for the scattering cross-section reduction due to multiple scattering effects.

B. Weak Imaging Regime (Tightly Confined Lattice)

• Setup: A Bose-Einstein Condensate (BEC) was loaded into a tightly confined 1D optical lattice (1064 nm) with a lattice spacing of 532 nm. The ground state wavefunction width was $\sigma_x \approx 21.2$ nm.
• Procedure: The authors used the weak imaging regime to selectively image a single lattice site (Site 0) by "cleaning" neighboring sites ($\pm 3$) using specific repumper frequencies and phases.
• Resolution: They successfully resolved a wavepacket with a standard deviation of 45 $\pm$ 5 nm.
• Significance: This is a factor of $\sim 4$ better than the diffraction limit and represents a direct measurement of a wavefunction narrower than the optical resolution. The slight broadening from the theoretical 21 nm was attributed to residual lattice misalignment angles ($\eta_y \approx 0.3^\circ$).

5. Significance and Impact

• Overcoming Diffraction Limits: This work provides a versatile, in-situ method to image ultracold atoms with nanometer-scale resolution, essential for studying strongly correlated phases in subwavelength lattices.
• Flexibility: The ability to switch between strong and weak regimes allows researchers to choose the optimal strategy based on the system's dynamics (e.g., fast dynamics require strong imaging; delicate, narrow states benefit from weak imaging).
• Scalability: The method relies on standard laser cooling and trapping techniques but adds a "dressing" step. It can be extended to other alkali metals (e.g., Cesium) and potentially to narrow-line transitions (e.g., Strontium, Ytterbium), though the criteria for strong/weak imaging would shift based on the transition linewidth.
• State Engineering: Beyond imaging, the technique allows for the spatial shaping and quenching of system dynamics, enabling the preparation of specific quantum states (e.g., single-site addressing) without destroying the rest of the cloud.

In conclusion, Veyron et al. have established a rigorous theoretical and experimental foundation for subwavelength microscopy of quantum gases, demonstrating that "dressed state" engineering offers a powerful, tunable pathway to bypass the optical diffraction limit in both fast and slow dynamical regimes.