The chemistry of AlF and CaF production in buffer gas sources

The question came from the molecular source

For laser cooling, a favorable electronic structure is not enough: the source must also produce a bright molecular beam with a phase-space distribution that downstream cooling and control can use. Experiments had shown a large contrast between AlF and CaF buffer-gas beams. The question was whether reaction chemistry inside the source could account for part of that difference and whether the precursor gas changes the outcome.

What the work did

Ab initio molecular dynamics was used to compare formation pathways for aluminum and calcium monofluoride after hot metal atoms encounter fluorine-donor gases in a cryogenic buffer-gas environment. The trajectory outcomes were then summarized with a tree-shaped Bayesian reaction model.

What it showed

The simulations predicted that NF₃ reacts more efficiently with the hot metal atoms than SF₆. Products formed with NF₃ also carry less kinetic energy and therefore require fewer helium collisions to thermalize. Across a broad temperature range, the calculated probability of forming AlF is substantially higher than that of forming CaF.

Experimental consequence

Reaction probability affects how many molecules are produced, while product kinetic energy affects how quickly they can cool through collisions with helium before leaving the cell. These are quantities that can be compared with source performance when choosing precursor chemistry or interpreting differences between molecular beams.

Relation to the AlF–AlF collision study

This work concerns formation from metal atoms and fluorine donors. The related AlF–AlF study starts after AlF has already formed and examines molecule–molecule collisions and transient complexes over 200–1000 K.

Related story: AlF–AlF collision dynamics: from the interaction landscape to complex lifetimes →