Ultracold experiments usually infer molecule-collision rates from how a prepared sample changes over time. Researchers trap atoms or molecules, let them interact for controlled hold times, measure the remaining population or reaction products, and fit those observations with a kinetic model that accounts for density and other loss processes. A disappearing molecule is evidence of loss—not, by itself, a count of every collision or an identification of what the collision produced.
What the experiment measures
The direct observation is typically a particle number, an internal-state population, a spectroscopic signal, or a detected reaction product. The rate coefficient is generally not read off an instrument: it is inferred by comparing measurements taken under controlled conditions with a model of the process.
As an Amazon Associate I earn from qualifying purchases.
For a two-body process, the number of possible interacting pairs depends on density. Researchers therefore need to characterize the sample’s density as well as its population decay. In a simplified, uniform sample dominated by two-body loss, a particle-number equation can be written as dN/dt = −K₂⟨n⟩N, where K₂ is the two-body rate coefficient and ⟨n⟩ is the density sampled by a particle. Trapped clouds may be spatially nonuniform, so analyses can require a density profile and a more detailed treatment than this illustrative equation.
How a population-decay measurement works
Prepare a defined sample
The experiment begins by cooling and confining selected species in specified internal states. The preparation matters: species, state, temperature, density, geometry, and dimensionality can all affect the inferred rate. A coefficient therefore describes the prepared system and the model used to analyze it, not a universal property of all ultracold molecules.
#1 Best Overall
For example, the 2022 KRb layer experiment began with degenerate potassium and rubidium atoms, loaded them into a one-dimensional optical lattice, and formed KRb molecules through magnetoassociation and stimulated Raman adiabatic passage. Microwave pulses and an electric-field gradient were then used to prepare selected rotational states and address individual layers.
Allow collisions for controlled times
Researchers hold the sample for a chosen duration, then repeat the preparation and measurement at multiple hold times. Comparing the resulting populations reveals how quickly particles disappear or move between states. Varying density can help distinguish a two-body contribution from losses that do not depend on the density of collision partners in the same way.
Measure remaining particles or products
Many experiments detect the atoms or molecules left after the hold, often by imaging. Some resolve particular states or layers rather than measuring only the total population. In the KRb study, researchers tracked molecule numbers in individual layers over time and fitted their decay to two-body loss equations.
Other experiments add product-sensitive detection, such as mass spectrometry or ion imaging. These measurements can reveal reaction outcomes or short-lived complexes that a reactant-loss curve alone cannot identify.
Fit the data to a kinetic model
The fitted coefficient depends on how the analysis treats the cloud’s spatial density, identical versus distinguishable collision partners, state mixtures, background loss, and coupled channels. The rate-equation convention also matters when comparing reported values. A sound result therefore pairs the number with its definition, preparation, and model rather than treating every reported “rate” as interchangeable.
What different measurement methods reveal
| Method | Primary observation | What it can establish | Important limit |
|---|---|---|---|
| Trap-loss or lifetime measurement | Remaining trapped population versus hold time | A loss rate or effective inelastic/reactive coefficient when density and competing losses are included in the model | Loss alone does not identify reaction products, and not every collision necessarily ejects particles. |
| Photoassociation spectroscopy | Resonances as laser frequency is scanned | Molecular levels and information useful for extracting scattering properties | It probes collision physics spectroscopically; it is not simply a count of lost molecules. |
| Product or complex detection | Detected reaction products or intermediate complexes | Reaction outcomes and information about short-range reaction dynamics | It answers a different question from a population-lifetime fit. |
| State- and layer-resolved measurement | Population decay resolved by internal state or spatial layer | How state preparation and geometry affect loss or exchange dynamics | The resulting rates belong to the particular state, layer arrangement, and model studied. |
Photoassociation is defined by Jones, Tiesinga, Lett, and Julienne as “the process where two colliding atoms absorb a photon to form an excited molecule.” Their 2006 review discusses how this spectroscopy can be used to study ultracold collisions. Product-detection approaches, reviewed by Liu and Ni in 2022, complement spectroscopy and loss measurements by providing information about products and complexes.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Published examples—and why their numbers are not universal
Cs–Cs₂ atom–molecule collisions
In a 2006 experiment by Staanum and colleagues, the Cs₂ storage time with Cs atoms present was reduced by a factor of seven compared with molecules stored without atoms, at a mean atomic density of 9.1 × 1010 cm−3. The authors reported atom–molecule rate coefficients on the order of 10−10 cm3/s. These are results for that Cs–Cs₂ system and its experimental conditions, not a general ultracold collision rate.
Layered KRb molecules
The 2022 KRb layer study reported fitted two-body loss coefficients of 2.99(17) × 10−3 s−1 for two rotational-state preparations and 1.78(24) × 10−3 s−1 for another, under the study’s stated configuration. It also found that rates varied with rotational state and adjacent-layer occupation. These fitted values describe that layered system and its analysis; they should not be compared directly with a coefficient from another system without checking definitions, units, density conventions, and conditions.
How to compare reported collision rates
Before comparing two published numbers, check whether the studies refer to the same physical quantity and comparable samples. In particular, compare:
- the colliding species and their internal states;
- temperature and collision-energy distribution;
- density and how it was calibrated;
- trap geometry and dimensionality;
- the measured observable—population loss, a spectroscopic signal, products, or complexes; and
- the rate-equation convention and which competing channels the fit includes.
A lifetime fit may yield an effective loss coefficient, while spectroscopy or product detection addresses different observables. Even when two papers use similar terminology, their values are not necessarily interchangeable unless these definitions and conditions align.
Quick Recap
References
- Jones, Tiesinga, Lett, and Julienne, “Photoassociation Spectroscopy of Ultracold Atoms and the Study of Physicists’ Molecules,” 2006.
- Liu and Ni, “Bimolecular Chemistry in the Ultracold Regime,” 2022.
- Tobias and colleagues, “Reactions between layers of molecules in a tilted lattice,” Science 375, 1299–1303, 2022.
- Staanum and colleagues, “Experimental investigation of ultracold atom-molecule collisions,” Physical Review Letters 96, 023201, 2006.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minute
