Determine the dissociative-recombination rate of PO⁺

Determine the dissociative-recombination rate coefficient of PO⁺, particularly under the extreme physical conditions of supernova-remnant shocks, to improve models of phosphorus chemistry and the PO⁺/PN abundance ratio.

Background

The chemical modelling identifies dissociative recombination with free electrons as the only destruction mechanism for PO⁺ in the adopted UMIST12 and KIDA networks. The rate coefficient currently used was introduced by Thorne et al. (1984) as an educated estimate rather than from a dedicated measurement or calculation.

Because reducing the adopted rate by a factor of ten substantially changes the predicted PO⁺/PN ratio, the uncertainty in this reaction directly affects the interpretation of the observed ion-dominated phosphorus chemistry in W44 and IC443. A more accurate rate is therefore needed for reliable modelling of phosphorus chemistry in highly ionised shock environments.

References

As already stated by , the adopted value is an educated guess and no dedicated studies are available to date. However, the abundances of molecular ions in highly-ionised environments such as SNRs can be strongly affected by the adopted dissociative recombination rates \citep[e.g.][]{priestley2017}. We find that already by reducing the \citet{thorne1984} value by a factor of ten, PO$+$/PN ratios greater than unity can be produced during the post-shock refreezing period ($\sim$10$4$ yr), as shown in Figure \ref{fig:lodr}. For comparison, previous shock modelling suggests characteristic interaction timescales larger than $104$ yr for W44 and $103$ yr for IC443 \citep[e.g.][]{cosentino2019,reach2019}. However, we note that these value are highly uncertain. Future efforts are needed to obtain a more accurate estimate of this reaction rate in studies of phosphorus chemistry, particularly in extreme environments such as SNR shocks.

— Chemical complexity in feedback from supernova remnants: first detection of PO+ in IC443 and W44  (2610.01211 - Cosentino et al., 1 Oct 2026) in Section 3.3, Comparison between models and observations