Cross-calibration of atomic pressure sensors and deviation from quantum diffractive collision universality for light particles

The room-temperature, velocity-averaged, total cross section for atom–atom and atom–molecule collisions can be approximated using a universal function depending only on the magnitude of the leading order dispersion coefficient, C 6. This feature of the total cross section together with the universal function for the energy distribution transferred by glancing angle collisions ( pQDU6 (Booth et al 2019 New J. Phys. 21 102001)) can be used to empirically determine the total collision cross section and realize a self-calibrating, vacuum pressure standard. This was previously validated for Rb+N2 and Rb+Rb collisions. However, the post-collision energy distribution is expected to deviate from pQDU6 in the limit of small C 6 and small reduced mass. Here we observe this deviation experimentally by performing a direct cross-species loss rate comparison for Rb+H2 and Li+H2 collisions. We measure a velocity averaged total collision cross section ratio of R=⟨σtotv⟩Li+H2:⟨σtotv⟩Rb+H2=0.83(5) . Based on an ab initio computation of ⟨σtotv⟩Li+H2= 3.104×10−15 m3 s−1, we deduce ⟨σtotv⟩Rb+H2=3.6(2)×10−15 m3 s−1, in agreement with a Rb+H2 ab initio value of ⟨σtotv⟩Rb+H2=3.574×10−15m3s−1 . By contrast, fitting the Rb+H2 loss rate as a function of trap depth to the universal function we find ⟨σtotv⟩Rb+H2= 5.52(9)×10−15 m3 s−1. This work demonstrates the utility of sensor-atom cross-calibration experiments to check the validity of theoretical computations to extend and enhance the performance of cold atom based pressure sensors.