Modeling sympathetic cooling of molecules by ultracold atoms

TL;DR

This paper models the sympathetic cooling process of ground-state CaF molecules by ultracold Li and Rb atoms, using differential cross sections and trajectory simulations to predict cooling efficiency and dynamics.

Contribution

It introduces a detailed simulation framework combining differential cross sections and trajectory calculations to analyze sympathetic cooling of molecules by ultracold atoms, highlighting Rb's effectiveness.

Findings

Rb is more effective than Li for cooling molecules.

The hard-sphere model accurately predicts molecule cooling rates.

Molecules can be cooled to 100μK in about 10 seconds.

Abstract

We model sympathetic cooling of ground-state CaF molecules by ultracold Li and Rb atoms. The molecules are moving in a microwave trap, while the atoms are trapped magnetically. We calculate the differential elastic cross sections for CaF-Li and CaF-Rb collisions, using model Lennard-Jones potentials adjusted to give typical values for the s-wave scattering length. Together with trajectory calculations, these differential cross sections are used to simulate the cooling of the molecules, the heating of the atoms, and the loss of atoms from the trap. We show that a hard-sphere collision model based on an energy-dependent momentum transport cross section accurately predicts the molecule cooling rate but underestimates the rates of atom heating and loss. Our simulations suggest that Rb is a more effective coolant than Li for ground-state molecules, and that the cooling dynamics are less sensitive to the exact value of the s-wave scattering length when Rb is used. Using realistic experimental parameters, we find that molecules can be sympathetically cooled to 100$\mu$K in about 10s. By applying evaporative cooling to the atoms, the cooling rate can be increased and the final temperature of the molecules can be reduced to 1$\mu$K within 30s.