🔬 atomic physics

Evolution of dipole-dipole dynamics in cold ammonia collisions

This paper reports the first experimental observation of state-to-state cross sections for cold ammonia collisions, revealing a counterintuitive suppression of dipole-dipole interactions at low energies due to the effective switching off of dipole moments, a phenomenon confirmed by scattering calculations and offering new avenues for controlling molecular collisions.

Original authors: Yao Chang, André J. A. van Roij, Stach E. J. Kuijpers, Sven Herbers, Etienne F. Walraven, Tijs Karman, Sebastiaan Y. T. van de Meerakker

Published 2026-07-28

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Yao Chang, André J. A. van Roij, Stach E. J. Kuijpers, Sven Herbers, Etienne F. Walraven, Tijs Karman, Sebastiaan Y. T. van de Meerakker

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 the invisible world of atoms and molecules as a bustling dance floor. For decades, physicists have been fascinated by "polar" molecules—tiny particles that act like microscopic bar magnets, with a positive end and a negative end. When these molecules bump into each other, they don't just bounce off like billiard balls; they feel a long-range tug-of-war, a "dipole-dipole" attraction that pulls them together. Scientists have long hoped to use these interactions to build quantum computers or create new states of matter, but to do that, they need to understand exactly how these molecules collide when they are moving very slowly, in the "ultracold" regime.

The big question has been: what happens when these magnetic dancers get really close? A classic theory, known as the Langevin model, predicted that as the molecules slow down, they would get pulled together more strongly, making collisions more frequent and dramatic. It was a simple, intuitive idea: slower speed means stronger attraction. However, quantum mechanics—the rulebook for the very small—suggests a much stranger reality. In the quantum world, a molecule's "magnetic personality" isn't always fixed; it can be hidden or "switched off" depending on its energy and how it spins. This led to a bold, counterintuitive prediction: at certain low speeds, the attraction might suddenly vanish, causing the molecules to ignore each other instead of colliding. Until now, this was just a theoretical ghost story, waiting for experimental proof.

This paper tells the story of how a team of researchers finally caught that ghost. They set out to test these predictions using ammonia molecules, the chemical cousins of the ones that make your nose twitch. The challenge was immense: to see these quantum effects, the molecules needed to be moving incredibly slowly, but standard lab equipment couldn't merge two beams of ammonia without them crashing into each other or flying apart. The team built a clever new machine, a "beam merger" that acts like a gentle, curved highway, guiding two streams of molecules to merge into a single lane where they can collide at speeds as low as 0.3 cm⁻¹.

What they found was a spectacular confirmation of the weird quantum world. As they slowed the molecules down, the collision rate didn't just keep climbing as the old theory predicted. Instead, it hit a "local maximum"—a peak—before suddenly dropping off. It was as if the molecules, upon realizing they were moving too slowly, decided to turn off their magnetic switches and stop pulling on each other. This "switching off" of the dipole-dipole interaction was the smoking gun the field had been waiting for. The researchers observed this peak for several different combinations of ammonia molecules, proving that this behavior is a universal rule for this class of particles, not just a fluke.

Even more fascinating was what happened below that peak. At the lowest energies, the researchers discovered that the molecules weren't just ignoring each other; they were interacting in a completely different way. Instead of the strong magnetic tug, a weaker, more subtle force called the "dipole-quadrupole" interaction took over. Using high-speed cameras (velocity map imaging), they watched the molecules bounce off one another and saw that the "dance steps" changed depending on which force was in charge. When the magnetic force was active, both molecules flipped their internal states; when the weaker force took over, only one flipped.

The study didn't just observe this; they mapped it out with incredible precision, comparing their real-world data with complex computer simulations. The match was so good that it confirmed the underlying math: the position of that mysterious peak depends on the specific energy gaps inside the molecules. The authors suggest that this discovery is a double-edged sword. On one hand, it means that trying to cool these molecules down further might be harder than expected, as the collisions that usually help cool them down suddenly become less effective. On the other hand, it opens up a new playground for control. Because these interactions are so sensitive, scientists might be able to use tiny electric fields to turn the magnetic attraction on and off at will, effectively toggling the molecules between "bouncing" and "sticking" modes. This paper turns a theoretical curiosity into a measured reality, showing us that in the ultracold world, the rules of attraction are far more playful and unpredictable than we ever imagined.

Technical Summary: Evolution of Dipole-Dipole Dynamics in Cold Ammonia Collisions

Problem Statement
The inherent collision properties of polar molecules remain largely unknown, particularly in the transition from high-temperature to ultracold regimes. While classical theories, such as the Langevin capture model, predict that collision cross-sections for dipolar molecules should increase monotonically as collision energy decreases (scaling as $E_{coll}^{-2/3}$ ), quantum mechanical descriptions suggest a more complex reality. In quantum systems, isolated molecules in non-degenerate states with definite parity possess zero dipole moment in the laboratory frame; a dipole moment only arises when states of opposite parity are mixed. Theoretical models predict that for molecules with parity doublets (near-degenerate states of opposite parity), the dipole-dipole interaction can effectively "switch off" at low collision energies. This leads to a counterintuitive local maximum (LM) in the scattering cross-section, followed by a suppression of the cross-section before the Wigner threshold law takes over at ultracold energies. Despite these predictions, experimental validation has been scarce due to the extreme difficulty in merging two beams of polar molecules with similar dipole moments and masses at sufficiently low energies to probe these regimes.

Methodology
The authors employed a novel experimental setup combining Stark deceleration, curved hexapole guiding, and a specialized beam merger to overcome the limitations of traditional merged-beam techniques.

Beam Preparation: State-selected packets of ammonia isotopologues ( $^{14}\text{ND}_3$ , $^{15}\text{ND}_3$ , $^{15}\text{NH}_3$ ) and nitric oxide (NO) were generated. The primary beam was velocity-tuned using a 2.6-meter Stark decelerator. The secondary beam was guided by a curved hexapole.
Beam Merging: A novel "merged guide" was utilized, where electric fields from six electrodes gradually evolved from two adjacent quadrupole traps into a single hexapole trap. This geometry allowed two near-copropagating beams to overlap with an effective crossing angle of $2^\circ$ in a near-zero electric field region, achieving minimum collision energies of $\sim 0.3 \text{ cm}^{-1}$ .
Detection: Scattered molecules undergoing the $|-\rangle \to |+\rangle$ inversion doublet transition were detected state-selectively using Velocity Map Imaging (VMI). To eliminate image blurring caused by ion recoil, a two-color laser ionization scheme involving vacuum ultraviolet (VUV) light was implemented. Isotope labeling ( $^{15}\text{N}$ vs. $^{14}\text{N}$ ) allowed for the spectroscopic separation of signals from the two beamlines.
Theoretical Framework: Experimental integral cross-sections (ICSs) were compared with coupled-channels (CC) scattering calculations based on ab initio potential energy surfaces (PES). Theoretical models were used to distinguish between dipole-dipole ( $|-\rangle \to |+\rangle$ coupling in both partners) and dipole-quadrupole ( $|-\rangle \to |+\rangle$ coupling in one partner) interaction mechanisms.

Key Results

Observation of the Local Maximum (LM): The study measured state-to-state cross-sections for $\text{ND}_3-\text{ND}_3$ , $\text{ND}_3-\text{NH}_3$ , and $\text{NH}_3-\text{NH}_3$ systems across an energy range of $0.3 - 100 \text{ cm}^{-1}$ . A distinct local maximum in the cross-section was observed for all systems, confirming the theoretical prediction. The position of the LM shifted to higher energies for systems with larger parity splitting energies ( $\Delta$ ).
Breakdown of the Langevin Model: At energies below the LM, the cross-sections deviated significantly from the $E_{coll}^{-2/3}$ scaling, showing a suppression of the dipole-dipole interaction strength. For $\text{ND}_3-\text{NH}_3$ , a plateau formed at energies below $\sim 1 \text{ cm}^{-1}$ , attributed to the dominance of the dipole-quadrupole interaction.
Universality of Scaling: Theoretical analysis revealed that the peak position of the LM ( $E_{peak}$ ) does not scale linearly with the total parity splitting $\Delta$ as previously suggested by simplified Langevin models. Instead, the data follows a power law: $E_{peak}/E_{dd} \propto (\Delta/E_{dd})^{4/3}$ , where $E_{dd}$ is the dipolar energy scale. This indicates a dependence on the reduced mass and dipole moments of the colliding partners.
Mechanism Identification via VMI: Differential cross-section measurements using VMI provided direct evidence of the interaction mechanisms. At low energies, the recoil energy distribution indicated that for $\text{ND}_3-\text{NH}_3$ , the undetected partner scattered elastically (dipole-quadrupole dominance), whereas for $\text{NO}-\text{NH}_3$ , a bimodal distribution suggested a mix of dipole-dipole and dipole-quadrupole interactions. This confirmed that the effective dipole moments are suppressed at low energies, preventing mutual polarization.

Significance
The paper claims that these results demonstrate how collisions between polar molecules evolve from the high-temperature limit to the ultracold regime in a counterintuitive manner, characterized by the effective switching off of dipole-dipole interactions at low energies. The findings have major consequences for the field:

Feasibility of Experiments: The suppression of inelastic cross-sections below the LM may impact the efficiency of evaporative cooling for trapped polar molecules, which relies on the ratio of inelastic to elastic cross-sections.
Interpretation of Data: Previous work and future experiments must account for the breakdown of the Langevin model and the presence of the LM when interpreting scattering data.
Control Opportunities: The sensitivity of the LM to external electric fields offers distinctive opportunities to control cold molecular collisions. By tuning external fields, it may be possible to switch effective dipole moments on or off, or toggle between dipole-dipole and dipole-quadrupole dominated interactions, providing new avenues for quantum simulation and computation.

More like this